.. -..- _ 'sa.

THESlS ’ ; .. . a...
if. L- i .-..- [.4 jg 1/ ;,

 
    
   

r mm " 'ty

This is to certify that the

thesis entitled

ANALYSIS OF ROUGH MILL FIELD STUDIES
presented by
Edward King Pepke
has been accepted towards fulfillment

of the requirements for

Ph.D. Forestry

 

 

degree in

 

Major professor

0-7639

 
  
    

91m:
25¢ per w per item

RETUMIM LIBRARY MATERIALS:

  

" “ fi‘“\\\ I; ~
{elm

y >~ ‘51-‘qu '4 Place in book return to raw
~. I “u. » a charge from circulation recon

v .

   

 

 

.w

to: o

ANALYSIS OF ROUGH MILL FIELD STUDIES

By
Edward King Pepke

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Forestry

1980

 

ABSTRACT
ANALYSIS OF ROUGH MILL FIELD STUDIES
By
Edward King Pepke

The analysis of ten rough mill field studies deter-
mined that the Optimum Furniture Cutting Computer Program
(OFCCP) provides a valuable service in assisting the wood
products manufacturing industry in procuring and processing
hardwood lumber. The cost and yield studies performed in
rough mills of furniture and wood dimension parts manufac-
turing companies resulted in sufficient information for
comparison to the OFCCP's predictions for yields, costs.
and grades to process. For all the data combined, the
companies' actual yields came to an average weighted by
lumber volume of 64.7 percent, 0.6 percent higher than the
computer predicted weighted average of 64.1 percent. If
the computer program's feature of customizing the yield
tables to the individual company's performance needed for
specific company applications, was incorporated the computer
could simulate both the exact yield and cost for a company.

The OFCCP follows the same basic steps a firm uses
when purchasing lumber. Since the steps are the same and

yield predictions are accurate,the program can be and is

being used by companies to purchase and process hardwood
lumber. The program will output the least cost grade,
yields, lumber volumes, costs,and other information for the
company's cutting bill piece requirements (part sizes and
quantities), and the associated costs of lumber, processing,
overhead,and other costs.

The diverse nature of the wood products industry's
products, production methods, and raw materials represented
in the ten sample companies used in this study precluded
the reporting of simple savings figures from use of the
OFCCP. The grade mix was examined and the processing ef-
ficiency was analyzed to determine if the OFCCP has any ef-
fect on lumber usage and rough mill costs. By improving
efficiency to the computer program's standards some companies
would save substantially--up to $250,000 in cost and 5A9
thousand board feet (MBF) annually. These are "potential"
savings and might never be fully realized by a company due
to limitations of plant design, capacity, labor efficiency,
lumber availability,and other factors. However, some of
the companies presently perform at or exceed the yield table
standards resulting in no savings through improved efficiency.

Similarly, the choice of grades to use (predicted
by the computer program) was compared to the grades actually
used by the company. The potential cost savings ranged from
$h9,000 to $2,523,000 and from 122 MBF to 1056 MBF annually

by using the optimum grade mix for the particular cutting

 

bill requirements and associated costs. Again, the high
value probably would never be reached due to various con-
straints affecting a specific company's operations. The
Opportunity for cost and lumber saings does exist in many
hardwood lumber processing companies through use of the

Optimum Furniture Cutting Computer Program.

ACKNOWLEDGEMENTS

The research for this doctoral dissertation began
in March 1977 under a graduate assistantship program jointly
sponsored by the Department of Forestry at Michigan State
University and the USDA Forest Service, Northeastern Area,
State and Private Forestry. While a student at Michigan
State University, the advice and teachings of all my pro-
fessors in forestry and the business management fields pro-
vided the basic foundation for my career in forest products
technology.

While employed by the USDA Forest Service's North-
eastern Area State and Private Forestry, my job duties
fortunately correlated well with the ten cost and yield
studies necessary for this dissertation. Since the infor-
mation obtained from the ten studies was needed by the For-
est Service, time was provided to perform research for and
write this dissertation. The author appreciates the en-
couragement and support offered by his employer towards
obtaining the Ph.D. degree.

The author also appreciates the assistance obtain
from his major professor and advisor Dr. Henry A. Huber.
The guidance offered by my professors and current committee
members Dr. Alan Sliker, Dr. Otto Suchsland, Dr. Stephen

Harsh, and Dr. Phillip Carter was essential to completion

ii

iii
of this dissertation and my doctoral degree. The support
of my family and friends in pursuit of my goals and inter-

ests was crucial to completion of this endeavor.

List of
List of
CHAPTER
CHAPTER
CHAPTER
CHAPTER
CHAPTER

CHAPTER

CHAPTER

TABLE OF CONTENTS

Tables. . . . . . . . . . . . .
Figures . . . . . . . . . . . . .

I Introduction. . . . . . . . . . . .
II Background . . . . . . . . . . . .
III Possible Sources ofNErrors. . .
IV Objectives . . . . . . . . . . .

V The Optimum Furniture Cutting

Computer Program. . . . . . . . .
VI Procedure. 0 O O O O O O O O O O 0

Manual Verification of the Optimum
Furniture Cutting Compter Program .

Field Testing of the Optimum Furni-
ture Cutting Computer Program . .

Analysis of the Data Generated From
A Rough Mill Field Study. . . . . .

VII Results and Analysis. . . . . . .
Analysis of the Optimum Furniture
Cutting Computer Program's
Yield Predictions . . . . . . . . .
Procurement of Lumber With Assist-
ance of the Optimum Furniture
Cutting Computer Program. . . . . .
Analysis of Grade Selection .

Analysis of Lumber Processing
EffiCienCyo I O O O O O O O O O O 0

PAGE
.iv

.vi

.28
.42

.42

.44

.52

-55

~55

.67
.71

.80

CHAPTER VIII Practical Roughi Mill Experiences

and Observations. . . . . . . . . . .90
Lumber Procurement. . . . . . . . . . .90
Lumber Processing . . . . . . . . . . .98
CHAPTER IX Summary and Conclusions . . . . . . . . 106
CHAPTER X Recommendations. . . . . . . . . . . . . 110
APPENDIX . . . . . . . . . . . . . . . . . . . . . 114

LIST OF REFERENCES . . . . . . . . . . . . . . . . 162

LIST OF TABLES
TABLE PAGE

3.1 Range of Cutting Bill Sizes By
Company and Lumber Grades. . . . . . . . . .23

5.1 Stacking Costs' Adjustment Factors For
The Optimum Furniture Cutting Computer
Program. 0 I O O O I O O O O O I O O O O O .34

5.2 Gluing Costs' Adjustment Factors For The
Optimum Furniture Cutting Computer Program .35

7.1 Summary of Each Company's Lumber
Yields by Grade. 0 o o o o o o o c o o o o .56

7.2 Summary of Actual Lumber Yields Observed
At Ten Rough Mill Field Studies. . . . . . .58

7.3 National Hardwood Lumber Association
Minimum Yields for Each Grade According
to Grading Rules . . . . . . . . . . . . . .59

7.4 Actual Average Yields by Volume of Lumber
Processed for Ten Study Companies
(Abstracted from Table 7.2). . . . . . . . .60

7.5 Summary of Optimum Furniture Cutting
Computer Program Predicted Yields For
Cutting Bills Processed at Ten Rough Mill
Field Studies. . . . . . . . . . . . . . . .61

7.6 Optimum Furniture Cutting Computer Program
Predicted Yields. Averages of Ten Study
Companies Weighted by Volume Processed
(Abstracted from Table 7.5). . . . . . . . .62

7.7 Plant Yield Adjustments for Ten Study
Companies by Grade . . . . . . . . . . . . .66

7.8 Maximum Yield Table Lengths From FPL 118

Vs. Maximum Lengths Actually Cut
During Study . . . . . . . . . . . . . . . .72

iv

TABLE ' PAGE

7.9 Potential Savings Through Optimum
Grade Utilization. I I I I I O I I I I 0 O I 76

7.10 Potential Volume Savings Through In-
creasing Efficiency to FPL 118 Yield
Levels I I I I I O I I I I I I I I I I I I I82

7.11 Potential Cost Savings Through In-
creasing Efficiency to FPL 118 Yield
Levels I I I I I I I I I I I I I I I I O I I83

7.12 Company One's Plant Yield Adjustments
By Grade I I I I I I I I I I I I I I I I I I84

7.13 Comparison of Actual Yields With and
Without Salvage Yields Included to the
Computer Predicted Yield Without Salvage
For Company One. . . . . . . . . . . . . . .85

7.14 Range of Potential Savings Through Ina
creased Efficiency. (Values from Tables
7010 and 7011) o o o o o c c o o o o o o o 089

8.1 Optimum Furniture Cutting Computer Program
Grade Choice Compared to Actual Grades Used
in Ten Field Study Rough Mills . . . . . . .94

8.2 Potential and Actual Production Rates for
Crosscut Saws and Ripsaws by Company and by
Grade of Lumber Processed in Ten Rough Mill
Field Studies . . . . . . . . . . . . . . 101

8.3 Increased Yield From Salvage in Percent By
Grade and by Company from Ten Rough Mill
Field Studies . . . . . . . . . . . . . . 105

9.1 Range of Potential Savings Incurred
Through Use of the Optimum Furniture
Cutting Computer Program at Ten Field
Study Rough Mills, 1977-1979. (Abstracted
from Tables 7.9, 7.10, and 7.11).. . . . . 108

LIST OF FIGURES
FIGURE PAGE

1.1 U.S. Producer Price Indexes Versus
Time (1970-1978) c o c o o c o o o o o o o .2

1.2 Example Yield Chart for 1 Common
From FPL 118. . . . . . . . . . . . . . . .5

1.3 Possible Rough Mill Layout for a Crosscut
Saw-First Sequence Rough Mill . . . . . . .6

5.1 Data Input to the Optimum Furniture
Cutting Computer Program . . . . . . . . 29

5.2 Blank Worksheet for the Optimum Furniture
Cutting Computer Program . . . . . . . . 30

5.3 Sample of Completed Worksheet for the
Optimum Furniture Cuttiner Computer
Program . . . . . . . . . . . . . . . . . 32

5.4 General Flowchart for the Optimum Furni-
ture Cutting Computer Program . . . . . . 37

5.5 Output Information from the Optimum
Furniture Cutting Computer Program. . . . 38

5.6 Optimum Furniture Cutting Compter Program
Sample Printout . . . . . . . . . . . . . 39

6.1 Flowchart of a Rough Mill Field Study . . 46

7.1 Comparison of Weighted Average Actual
Lumber Yields Observed at Ten Rough Mill
Field Studies (From Table 7.2) to Weighted
Average Optimum Furniture Cutting Computer
Program Predicted Lumber Yields for Cutting
Bills Processed During the Same Field
Studies (From Table 7.5). . . . . . . . . 63

vi

CHAPTER I

INTRODUCTION

Hardwood lumber users compose the core of the fur-
niture and dimension (semi-finished wood parts of specific
size for remanufacture) industry in the United States. His-
torically, the supply of high quality lumber for the industry
has been readily available at relatively low prices; but in
the last few decades, the extensive use of hardwood timber
for furniture, pallets, railnxad ties, and other purposes
has seriously depleted the current resources. Conversion of
prime hardwood timber growing land in the United States to
pine (softwood) plantations, agriculture, utility and trans-
portation right-of—ways, and urban and industrial develop-
ment sites threatens the present and future of quality hard-
woods. Even with heightened interest in growing quality
hardwood timber and in improving management of the present
hardwood forests, unfavorable economic conditions will con-
tinue to plague lumber consumers.

The economics facing the manufacturer employing
hardwood lumber as a basic raw material can best be de-
scribed graphically. Figure l.1, "Producer Price Indexes
vs. Time", illustrates that the producer price indexes

(formerly "wholesale price index") for hardwood lumber,

230'
220‘
2101

200‘

   
    

Wood
Office
Fuel Hardwood Furniture
1801 Lumber a

1404

Producer Price Index (1967 = 100)

l4 I4
N \n
C) C)

 

....IIIIIII
I

 

T Y

1970 '71 '%2 '%3 '51' '75 '76 '57 '%8

Year
FIGURE 1.1 U.S. PRODUCER PRICE INDEXES VERSUS TIME (1970-1978)

Source: U.S. Bureau of Labor Statistics, Producer Prices and

Price Indexes Supplements. 1970 to 1979, Washington,
DICI

U.S. Bureau of Labor Statistics. Handbook of Labor

Statistics 1978, June 1979, Bulletin 2000, Washington,
DICI

3

labor, and fuel are rising much faster than the index of
furniture. The manufacturer is financially squeezed when
his costs rise faster than the price obtainable for his
product. Thirty-five to forty percent of the cost of fur-
niture and fifty percent or more of the cost of dimension
parts production are due to lumber alone.l Cost reduction
has become essential; and, one of the most lucrative areas
for savings is in the cost of processing hardwood lumber.

The majority of logs in the hardwood growing stock
are grades 3 and 4, the poorest logs for lumber production.
Log grades 1 and 2 are the primary grades converted by the
sawmill into lumber. From log grade 1, the average yield
for the species used by the ten companies in this study came
to 32.5 percent FAS ("FAS" is the commonly used abbreviation
for "firsts and seconds?the highest lumber grade), 9.2 per-
cent select, 30.0 percent 1 common, 12.7 percent 2 common,
and 16.8 percent 3 common (3 common is the lowest yielding
lumber grade normally used for furniture and dimension pro-
duction). For log grade 2 the quantities of FAS, select, 1,
2 and 3 common are 7.7, 5.0, 34.0, 23.5, and 29.9 percent
respectively.2

The OFCCP was developed by the USDA Forest Service
and Michigan State University to improve the utilization
efficiency of hardwood lumber. The specific purpose of the
computer program is to enable hardwood lumber users to make
purchase decisions using a scientific method for analyzing

available grades and species, respective costs, and the firm's

4
product requirements. The data base comes from the Forest
Products Laboratory publication, Charts for Calculating
Yields from Hard Maple Lumber, FPL 118.3 See Figure 1.2
“Example Yield Chart for 1 Common Lumber Grade from FPL 118."
Black walnut and red alder yield tables may be accessed
through the computer program, but were not needed for this
research. The program's two basic functions are to predict
(1) the best grade mix and (2) the lumber yield for a manu-
facturer's cutting bill with its associated costs.

The OFCCP's predictions were checked against manually
calculated yields using the published yield charts. A satis-
factory comparison of manual and computer predictions led to
the introduction of the computer program in the hardwood lum-
ber using industry in Michigan and other states. Favorable
user feedback inspired more detailed confirmation of the com~
puter program's ability to accurately predict lumber yields,
procurement information, and cost predictions.

A series of ten yield and cost studies were performed
according to a procedure adaptable to the widely varying con-
ditions encountered at the various companies. Basically,
the tests involved monitoring the volume of rough lumber
going into a rough mill (a place where rough, unsurfaced,
variable-sized lumber is machined to fairly uniform "rough"
dimensions). See Figure 1.3 "Rough Mill Layout". The pro-
cessing of the lumber was observed and timed, and the final

net product output was measured and recorded.

  
 

7 J

I ‘
E \FQHQka Q uh: Qhk bxfihkk

~hhtfi§§ Wu§i k 8 KO r: Qt.
8‘ - k U‘Qk fix“

ADJUSTMENT FOR WIDTHS OTHER THAN 2 INCHES

 

FIGURE 1.2 EXAMPLE YIELD CHART FOR 1 COMMON FROM FPL 1183

.SOmomo .pcmgnom .mwm 2H :awoaasm ..o:H .ohooEISOvacf/hH “condom

.AHHE IUDom MOZMDomm EmmHm13<m abommomo < mom BDOM<H QAHS :Ubom mqummom m.H mmDUHm

U0:

mUZHUQm mom 3<m mw<>q<m
mflm<ezmbe

AOEBV m3<m mu<>q<m
Am>Hmv m3<m mHm

H Acmezoo Mmosms :BHS mmemom eoqm chsz
AOBBV m3<m UZHUQM

mmz<gm

Ammmzev m3<m mmOIBDo

mmmo<BmZD UZHBHHB

\OL\COO\O

 

 

 

r—INMd'W

 

7

The volume and time data collected at ten companies
were used to calculate the individual firm's costs and yields.
Using the OFCCP each company's actual study results were sim-
ulated. The computer predictions were compared to the actual
study results to determine if the computer program success-
fully accomplishes this research's objectives (listed in

Chapter III).

CHAPTER II

BACKGROUND

The purpose of this chapter is to acquaint the reader
with the history and background of research leading to the
creation of the Optimum Furniture Cutting Computer Program
(OFCCP). The background section will discuss some of the
early efficiency pioneers in the hardwood lumber using in-
dustry who had foresight to recognize the need before the
more popular but wasteful short term sentiment that the qual-
ity wood resources would be ever bountiful. Those farsighted
men realized that efficient procurement and processing of
hardwood lumber could have alleviated the present shortage
of the high quality wood resource. While the economic prin-
ciples of price versus quantity prevail in the free enter-
prise system of the United States sound ecological conse-
quences will result when hardwood lumber is used most ef-
ficiently; reduced waste will lead to extension of the natural
resource. While no one chapter could do proper justice to
all the people involved in the improvement of hardwood lum-
ber utilization, this chapter will highlight the pertinent

developmentary steps forming the foundation of the OFCCP.

9

One man who initiated the momentum in the early
1950's which led to the current interest in lumber use ef-
ficiency was C.D. Dosker of Gamble Brothers Incorporated, a
wood dimension parts manufacturing company located in Louis-
ville, Kentucky. Besides realizing the economic advantages
in efficiently procuring and processing hardwood lumber he
visualized the whole scene. Dosker led the industry into a
consciousness of the negative consequences resulting from
wasteful wood usage. It was directly from Dosker's prodding
on the urgency and importance of increasing lumber use ef-
ficiency that the USDA Forest Products Laboratory began
fundamental research on lumber yields--the same research be-
came the data base for the OFCCP.

Another man who worked closely with Dosker at Gamble
Brothers was H.C.'Moser. Moser's views are well documented
in an early paper presented at a North Carolina State Uni-
versity symposium on lumber yield titled FPotentials for
Increased Profits Through Analysis of Yields and Operational
Procedures".# Moser put the situation into an illuminating
perspective when stating the average ratio of lumber cost
to labor cost in the rough mill is 10 to 1. Since this ratio
is still extremely lopsided towards lumber the other ideas
promoted by Moser maintain validity. Moser's ideas on yield
and cost fit well with Dosker's in planning a comprehensive
program to save lumber through efficient operations through-

out a company. The paper concludes with this statement:

10

There is a danger that sacrifices of

yield to the gods of automation may be

false religion. Not only that, it can

be poor business.

Along with Moser, Vincent R. Ross, President of Ross
Associates a consulting firm headquartered in Ashville, North
Carolina, participated in the North Carolina State University
Rough Mill Seminars. Ross has unselfishly given of himself
in the quest to increase the efficient utilization of hard-
wood lumber. Ross was a key figure in the movement to make
peOple in the industry aware of thesignificance of profitably
improving lumber yield. While Ross has many presentations
and publications to his credit, the one presented at the
December 1966 Rough Mill Yield and Operations Seminar en-
titled "Management's Role in a Program for Yield Optimiza-
tion" demonstrates his view of the overall picture;3 Ross'
point, very true today also, was that without full manage-
ment support and involvement the success of a yield improve-
ment program will be seriously compromised at best.

At the same time period that North Carolina State
University was conducting seminars on rough milling it was
also performing basic research on hardwood lumber yields.
R.J. Thomas published three volumes, with the first in 1956
on the "Number of Maximum Yield Cuttings per 1000 Board
Feet."6 The data collection was performed on graded yellow
pOplar lumber. Unfortunately the North Carolina Yield Tables
were cumbersome to use for the commercial firms needing the

information most. State extension specialists like W.E.

ll

Keppler and Steven Hanover were instrumental in initiating
their acceptance along with industry consultants like Vincent
Ross and Ed Shock. The widespread use of these early yield
tables was to come later when the information would be com-
puterized by Sandy Mullin in the Barr-Mullin Company, pro-
ducers of the Mini-Max lumber yield improvement computers.

Thomas also worked with Keppler to push the use of
the lumber yield information by means of open rough mill

7

seminars, company demonstrations, and trade journal articles.

The questions posed in two of Keppler and Thomas' articles

in Wood and Wood Products in 1965 entitled "New Two-Way Sys-

tem Predicts Lumber Yield"8 and fiHow to Predict Costs by

9

Using New Yield Data" are valid today and indicate the need

for the OFCCP:

"1. What utilization or yield can be expected
when a certain grade of lumber is used to
produce a given number of rough cuttings
of various dimensions?

2. If a special type of cutting must be pro-
duced how does this affect overall yield?

3. What is the Optimum combination of grades
to use to produce a required number of
different cuttings?

4. How shall stock of lumber be allocated to
various products?

5. What are reasonable costs for such al-
locations?"

The articles by Keppler and Thomas explained how the North
Carolina State University yield tables could solve these

problems. By special type of cutting in question 2 above,

12
Keppler and Thomas were referring to an extraordinary length,
width, and/or quantity requirement which might adversely af-
fect overall yield.

Dosker's impetus led the USDA Forest Service into de-
veloping yield tableS‘MJprovide similar information as the
North Carolina State University research. Much of the orig-
inal work was performed at the Forest Products Laboratory in
Madison, Wisconsin and the Forestry Sciences Laboratory in
Carbondale, Illinois. The research performed led to many
publications by such people as Daniel E. Dunmire and George
H. Englerth in their publication VProgramming for Lumber
Yieldn";O In the introduction of the above Forest Products
Journal article Dunmire and Englerth set the stage with
these words:

To a considerable extent this project was

developed and conducted as a result of the

interest displayed by Dosker. C.D. Dosker,

for many years a leader in the hardwood

industry, stated the problem thus: 'Wood

is missing out as a raw material in thousands

of usages simply because the industry has no

information on which a designer, or a user, can

determine in advance what the refining costs
of wood as a raw material will be'.

Dunmire and Englerth's article and procedure for the
computer prediction of lumber yield drew on some basic com-
puter programming performed in the early 1960's. At the
Forest Products Laboratory a computer program was designed
to determine the dimension and number of cuttings obtained
from a board whose size and defects had been entered into

the computer. The program was detailed in a publication by

13
Claudia Wodzinski and Eldona Hahm titled A Computer Program
to Determine Yields of Lumber.ll Their computer program had
flexibility and became the basic tool in development of yield
charts.

A year later, 1967, Dunmire and Englerth published
another Research Paper (NC-15) Development of a Computer
Method for Predicting Lumber Cutting Yields.12 At this same
time David R. Schumann in conjunction with the other above
named authors at the Forest Products Laboratory published
some of the first functional yield tables for hard maple

13’l4 (and therefore all similarily graded species of

lumber
lumber), black walnut,15 and alder.’16 The black walnut
charts were a more refined update of research began in the
early 1950's by the American Walnut Manufacturers Associa-
tion in their publication "Walnut Yields".l7 Basic data
collection and research for black walnut was performed at
this same time by J.W. Creighton, W.G. Stump, and W.F.
Hutchins in their 1952 publication "Correlation of Walnut
Furniture Cutting Requirements with Grade Yield." 18

With the basic foundation laid in the above cited
publications, the next step was verification of the data
and demonstration of its application to the wood using in-
dustry in order to achieve the original goals of efficient
utilization of the hardwood resource. Various pOpular trade
journal articles were devoted to explanation of the useful-

ness of the yield table predictions. One such article ap-

peared in the March 1968 Wood and Wood Products titled "New

14

Yield Data for Cutting Hardwood Dimension Stock".‘19 Con-
firmation<xfyield table accuracy was performed in numerous
studies like Henry A. Huber and David R. Schumann's "Com-
parison of Predicted and Actual Yields of Dimension in the
Grand Rapids, Michigan Area".20 Huber and George Vasiliou,
Extension Specialists in Forest Products at Michigan State
University's Forest Products Department performed and pub-
lished on other cost and yield studies in "Cost Analysis in
Wood Products Manufacturing".21 Later Huber again used the
Forest Products Laboratory data and computer program in his
publication "In the Rough Mill Should You Rip or Crosscut
First?".22

Attempts to increase processing efficiency continued
with Forest Products Laboratory developed yield tables in
the 1960's and early 1970's but the need for a more compre-
hensive computer program was apparent. Robert Nevel and
associates at the USDA Forest Service's Princeton, West
Virginia Forest Products Marketing Laboratory developed
OPTIGRAMI. The nomographs from Englerth's FPL 118 Research
Paper Charts for Calculatinngimension Yield for Hard Maple
Lumber were the basis of OPTIGRAMI once they had been com-
puterized. The yield information was combined with a linear
program to predict the OPTImum GRade MIx (hence OPTIGRAMI)
for any given cutting bill (a parts list of specific quan-
tities and dimensions to be produced) and the associated
processing and procurement costs. OPTIGRAMI is designed to

use punched card data input which makes the program unwieldly

15
to use on site in the field. At the time of this writing
OPTIGRAMI has not been published although much of the work
has been completed.

To overcome some of the limitations of OPTIGRAMI
the USDA Forest Service, Northeastern Area State and Private
Forestry, entered into a COOperative agreement with Michigan
State University's Department of Forestry through Professor
Henry Huber. Stephen Harsh, Professor of Agricultural Eco-
nomics at Michigan State University, performed most of the
computer programming for what was to become the fOptimum
Furniture Cutting Computer Programf. The computerized yield
data of FPL 118 Charts for Calgggating Dimension Yield for
Hard Maple Lumber from the OPTIGRAMI program was combined
with a cost minimization linear program. The resulting pro-
gram was designed to be accessable by remote data terminal
linked to the main computer over ordinary telephone lines.
The ease of access enabled the OFCCP to be used immediately
within a company any time information is needed.

Validation of the OFCCP was performed by manually
calculating cutting bills through the FPL 118 hard maple
nomographs and comparing the results to the computer pre-
dicted yields at Michigan State University. Further checking
of the functionality of the computer program was done through
a series of demonstrations in Michigan. Results were favor-
able in that computer predictions came close to actual in—

dustry experience.

 

 

16

The USDA Forest Service realized the imposing neces-
sities of using the hardwood resource to its maximum ef-
ficiency. Through the efforts of many of the forenamed men
the Northeastern Area, State and Private Forestry branch of
the Service established RIP, the Rough-mill Improvement Pro-
gram. The author was hired in June 1977 to implement savings
in the 20 state Northeastern Area and coordinate efforts in
the remaining states to improve the utilization standards of.
the hardwood resources.

Most of the emphasis of the RIP program has been to
help the secondary (in this case lumber to semi-finished or
finished goods) wood manufacturing industry improve lumber
yields thereby reducing wood waste and reducing costs. The
OFCCP has been the main tool used in conduction the RIP
work. Various publications aimed at the industry proces-
sing hardwood lumber have appeard in trade journals.23’2u'25’26

As a.result of RIP the OFCCP has been used to help
over 200 companies in the United States and a few outside
the United States.

Trade journal articles plus shorter press releases
for trade journals and trade associations new letters re-
sulted in more demands for on site demonstration by individ-
ual companies and by secondary processing industry groups.
Since the computer program was developed it has been pre-
sented to: Grand Rapids Furniture Manufacturer's Association,

National and Regional Forest Products Research Society

17
meetings, National Association of Furniture Manufacturers,
Southern Furniture Manufacturers Association, Hardwood Di-
mension Manufacturers Association, Keystone Kiln Drying As-
sociation, New England Kiln Drying Association, Northeastern
Dimension Manufacturers Association, and various school
groups. Typically each presentation would result in a few
invitations for in-plant demonstrations at individual furni-
ture and dimension manufacturing companies.

Each presentation or demonstration was performed with
the knowledge of or actually with local and regional forest
products utilization specialists involved. The utilization
specialists were federal, state or extension specialists or
industry consultants. These people often carried the program
to their constituents. In the past three years the versa-
tility and many applications (from shoe heels, stairs and
piano keys to the most eXpensive high quality wood furniture)
of the RIP program has been of assistance to the industry.
The necessity of in-depth cost and yield studies at some manu-
facturing plants enabled the gathering of data for the ten

scientific tests described in the next chapters.

CHAPTER III

POSSIBLE SOURCES OF ERROR

While it was the intent of this research to be as
objective and unbiased as possible, certain sources of er-
rors were unavoidable. The expressed intent of the cost and
yield studies was to monitor each company's "normal" oper-
ations. Unfortunately. some uncontrolable variables were
present: since these potential sources of error could af-
fect the study results, an explanation is required.

When an observer watches employees engaged in their
routine tasks, his presence may influence a worker to alter
the rate and the method in which the job is done. To elim-
inate or alleviate this source of error in the present study,
the employees and foremen were briefed in advance about the
purpose of the observers. To minimize the possibility that
workers might alter their production rates, they were in-
formed that the study staff was only measuring the capacity
of the machinery, not the worker's individual production rates.
However, since the workers control the rough mill machinery
(not vice—versa), the explanation of the study purpose was

misleading: in actuality, the workers' production rates were

18

 

19

being measured. At some companies, employees either ac-
cepted the explanation that the machinery's capacity was
being measured, or they did not care enough to alter their
normal production rates. In other companies, a deviation
from standard work rates was observed during the first portion
of the study.

If workers thought the research team was establish-
ing labor rate standards for the company, the rough mill
workers might have tried to intentionally slow their speed
to set a low, easily exceeded, production rate. On the
other hand, some workers might have tried to increase their
work rate to obtain a favorable report with supervisors.

Several times during a study, when the researcher
suspected the workers were performing at rates above or be-
low the company's normal production level, the supervisory
personnel were questioned about relative production rates.
In some instances, the supervisor confirmed that a worker
was performing faster or slower than normal. But in every
case when a change in usual work rates was initially iden-
tified, management later confirmed that the workers had
slipped back to more typical rates. If time to perform a
study was unlimited, the "study effect" could have been
reduced by disregarding the initial data collected, favor-
ing the less biased data which followed. The potential er-
ror caused by changes in normal methods and in normal rates
probably did not significantly affect the ten studies. The

researchers' initial concern for changes in the normal tasks

20
performed due to observers' presence was not confirmed by
supervisory personnel.

Another source of error introduced by the study pro-
cedure was the change from a company's typically mixed grade
packages to distinctly graded lumber packages. In order to
scientifically monitor yield and production rates for each
grade, the company was requested to have a National Hardwood
Lumber Association Inspector grade and measure the volume of
the study's lumber. None of the ten companies normally seg-
regated all of their lumber into distinctly graded packages
before processing it through the rough mills: several com-
panies did purchase lumber grade mixtures oriented towards
specific grades, either high or low, for their desired cut-
ting schedules.

The grades were physically separated before the start
of the study to preclude disruption of the rough mill routine.
Usually processing of distinct grade packages went uninter-
rupted because it was possible to closely monitor the time
when the next grade would be processed and to record the
relevant volume measurements. In several plants, however,
it was impossible to make accurate measurements of each grade
due to a continuous production line linking the ripsaws and
crosscut saws: aCcurate measurement required completely
purging the rough mill line of one grade before beginning
to process the next grade. These alterations of normal pro-
duction practices were effectively eliminated by not re-

cording the time incurred in change of grades for study

21
purposes. The researcher was able to distinguish between
time delays due to a change of grade to meet the study pur-
poses and routine time delays due to bringing more lumber
into production when the supply had been depleted. To avoid
any influence on the measured production rates, the delays
due to conducting the study were not included.

Processing individual grades was a possible source
of error when the cutting bill lengths were not arranged by
grade. For example, a company may have normally processed
mixed grade cutting bills having a variety of lengths, from
long to short, and a variety of widths, from wide to narrow,
from a mixture of higher grades, 1 common, select, and FAS;
low yields could result when the company attempted to cut
the same length and width pieces from a single grade at a
time. To alleviate this source of abnormally low yields,
the companies were requested to process their own typical
cutting bills during the study. Since the target volume
for each field test (20,000 board feet) occasionally ex-
ceeded the volume of one cutting bill's lumber requirements,
several cutting bills were combined to utilize all the study's
lumber. When combination cutting bills were processed, they
were treated as a single cutting bill by the computer sim-
ulation. 1

The companies processed cutting bills in the ac-
customed manner. Introducing individual grades presented

a problem to some rough mills when they had no available

22
knowledge of the most efficient means of scheduling the ap-
propriate cuttings from the correct grade. The researchers
abstained from assisting the firms to determine which cute
tings to produce from each grade. Since most companies used
their normal cutting bill from their normally available grades,
this presented no large source of error. In other cases, the
loss in yield due to cutting lengths not matched to the grade
being processed was evident. While the effect on yield due
to grade segregation was observed to be minimal, the net re-
sult cannot be accurately estimated. The magnitude of this
source of error was not measured, but its potential presence
must be acknowledged.

The effect on yield due to a company's cutting bill
requirements has been mentioned in Chapter II. Accurate
matching of the cutting bill's lengths, widths, and piece
requirements is essential to obtaining maximum cutting yields.
Since the ten field study company's yields will undoubtably
be compared, the reader is cautioned that the cutting bills
produced and the lumber used were not the same for each com-
pany. When a company processes high grades of lumber it will
obtain better yields for a given cutting bill than when it
uses a lower grade. The cutting sizes and the grades pro-

cessed by each company are listed in Table 3.1 on the next

page.

 

 

 

23

TABLE 3.1 RANGE OF CUTTING BILL SIZES BY COMPANY AND
LUMBER GRADES

 

 

 

 

Lumber Range of Length 'Range of Widths
Com- Grade Minimum Maximum Minimum Maximum Random Width
pany Processed (inches) (incneg) (inches) (inches) for Glued Panels

1 FAS 13 43 N/A N/A yes

Select 13 52 N/A N/A yes

1 Common 13 48 N/A N/A yes

2 Common 13 52 N/A N/A yes

3 Common 13 43 N/A N/A yes

2 FAS+Select 13 66 l 6 no

1 Common 13 66 l 6. no

2 Common 13 - 64 1 63 no

3 PAS 11 81 1 3/4 1 3/4 no

Select ‘11 81 1 3/4 1 3/4 no

1 Common 11 81 1 3/4 1 3/4 no

2 Common 11 81 1 3/4 1 3/4 no

4 FAS+SeleCt 14 101 Zé 5 no
1 Common 13 96 2 5 no

2 Common 13 4O 2 2% no

3A Common 13 28 2% 2% no
5 ’ PAS 15 80 N/A N/A yes

Select 15 61 N/A N/A yes

1 Common 15 56 N/A N/A yes

2 Common 15 56 N/A N/A yes

3 Common 15 56 N/A N/A yes

6 FAS+Select 18 58 N/A N/A yes

1 Common 18 ' 58 N/A N/A yes

2 Common 18 58 N/A N/A yes

7 1 Common 17 52 2} 2 no
2 Common 17 52 2% 2 no
8 Select 17 81 1 5/8 3% no
1 Common 17 81 1 5/8 3 no
2 Common 20 81 1 5/8 3: no
9 PAS 14 71 1* 5 no

Select 15 71 1 3 no

1 Common 13 60 21 21 no

2 Common 13 6O 2. 2. no
10 1 Common 12 80 N/A N/A yes

2 Common 12 35 N/A N/A yes

3 Common 12 36 N/A N/A yes

N/A 8 Not Applicable (No specific minimum or maximum width)

Source: Ten rough mill field studies performed 1977-1979 in
the Eastern United States.

24

Another uncontrollable variable was a company's
production sequence, i.e., whether the firm crosscut or
ripped the lumber first. Theoretically, the companies pro-
cess the lumber in such a way to maximize product yield.

For example, flooring manufacturers ripsaw lumber before
crosscutting to length because their product is long and
narrow. Conversely, furniture cabinetry manufacturers
initially crosscut incoming lumber to length to obtain the
wide widths necessary to meet their product requirements.

The computer's data bank, derived from FPL 118 is
based on crosscut first lumber processing. The yield tables
are being successfully used by a number of rip-first lum-
ber processers; however, different production sequences
could affect the study results. To minimize variability
from this source, yield summaries for companies with dif-
ferent production sequences will show results for each in-
dividual sequence before combining them together.

While lumber grading can be mathmatically eXplained,
human judgement will always be involved because wood is a
heterogenius substance. Since an inspector's subjective
judgement is involved, some margin of error is possible.

A fairly well accepted standard is that any given lot of
lumber may be within five percent of the designated grade.
Gordon Bullard, Chief Lumber Inspector of the National Hard-
wood Lumber Association, confirmed that a five percent range

was "a pretty fair yardstick".28 Bullard stated that grade

25
variance will occur because profit demands rapid lumber in-
spection and inspector judgement varies about the extent of
knots, wane, stain, surface checks, and honeycomb.
Because of potential legal consequences, the "five

percent rule" has never been documented; however, Rules for

 

the Measurement and Inspection of Hardwood and Cypress Lum-

27

ber makes a provision for deviation in grade called "the
four percent money clause." This clause is used to deter-
mine whether or not a discrepancy in grade exists when a
lumber buyer believes the quality of lumber received differs
significantly from the quality listed on the invoice. When
a buyer requests a formal inspection of the lumber by the
National Hardwood Lumber Association, the four percent money
clause is used. The clause is documented in the following
passage from the National Hardwood Sales Code, Article X ~-
Inspection Section 5:

Should this original official inspection

result in not more than 4% deductable dif-

ference in money value from the gross amount

of the invoice, the buyer is to pay all ex-

penses of the inspection, accept all lumber

and honor the seller's invoice in full. If

the deductable difference be more than 4%

money value, the seller is to pay all ex-

penses of the NHLA inspection and labor

charges at actual cost or at the rate of

$4.00 per M feet, whichever is less, the

seller to issue an invoice for the items
reported on the inspection certificate.29

The clause is also used as a standard during reinspections
as illustrated in this portion of the National Reinspections

'Regulations and Guarantee:

 

 

26

V. If the reinspection results in a

difference in favor of the party com-

plaining, of more than four percent in

money value based on the total gross

value of all stock included in the in-

spection, the party complaining may re-

ceive the amount of such difference

directly from the Association by sending

to the secretary an itemized statement

showing in detail the items and amounts

as shown on the original certificate and

reinspection certificate.30

The intention of this chapter was to investigate
potential sources of error which might influence results
of the ten rough mill studies. While the researcher at-
tempted to minimize error in the study, he was unable to
totally control the ten companies operating in a compet-
itive business environment. Regardless of the long range
benefits resulting from the cost and yield studies per-
formed at a company's plant, higher priority is given to
short range obligations (production schedules, shipment
deadlines, profits, and others). No firm can forego its
obligations in order to participate in a study. The ten
studies were performed in as controlled an environment as
possible to minimize the possible sources of error. How-
ever, the reader should be aware of these sources of var-

iation and should interpret the results of these rough mill

studies accordingly.

CHAPTER IV

OBJECTIVES

The objectives of this research are:

1.

To analyze the capability of the Optimum Furni-
ture Cutting Computer Program for predicting
lumber yields, given information on lumber grade
requirements, associated costs, and cutting bill
needs.

To determine if the Optimum Furniture Cutting
Computer Program can be used by the wood pro-
ducts manufacturing industry for the purchasing
of lumber on a scientific, factual basis.

To determine whether or not the opportunity for
monetary and lumber savings exists in the sec-
ondary wood processing industry through use of

the Optimum Furniture Cutting Computer Program.

27

CHAPTER V

THE OPTIMUM FURNITURE CUTTING COMPUTER PROGRAM

The Optimum Furniture Cutting Computer Program
(OFCCP) was developed by the USDA Forest Service and
Michigan State University to increase efficiency in the
utilization of hardwood lumber. The program was specifi-
cally designed to enable purchasing and processing of lum-
ber on a scientific basis while taking into account the
available grades and species, with their respective costs,
in conjunction with a firm's product requirements.

The computer program contains two main steps: the
first being an analysis of the maximum yield for the firms
cutting bill requirements for each availablegrade. The
second step combines the relative costs of lumber, labor,
and overhead for various production steps (preceeding and
including the rough mill operation) with yield information
in a cost minimization linear program.

The inputs to the computer program are listed in
Figure 5.1, "Data Inputs to the Optimum Furniture Cutting
Computer Program". These values are usually written on a
blankWCrksheet like the one in Figure 5.2. Figure 5.3 is

an example of a completed worksheet ready for computer entry.

28

 

 

 

 

29

Lumber Species

Lumber Thickness

Delivery Cost

Drying Cost

Percent Drying Shrinkage
Interest Rate

Average Inventory Period
Stacking & Handling Cost
Gluing Cost

Lumber Cost by Grade

Processing and Overhead Cost by Grade
Plant Efficiency Level by Grade
Cutting Bill

Salvage Lumber Value

Salvage Lumber Size

Salvage Lumber Usage

Other Costs

FIGURE 5.1 DATA INPUT TO THE OPTIMUM FURNITURE CUTTING
COMPUTER PROGRAM.

 

30

WORKSHEET

Optimum Furniture Cutting Computer Program

_;

1

Iv

 

 

 

 

 

Date:
Bill 0:
File 5:
GENERAL AND COMM)” COST INFORMATION
'Scare Enter all data with
Company 01 '"" a"-—"underneath.
Species ._.....
Thickness in 11473 . . . . . . . . . . . . . . . . . . .02 a...
“live” cost in slflF I I I I I I I I I I I I I I I a“-

Owing cost 1“ SITE? I I I I I I I I I I I I I I I
shrinkage in z I I I I I I I I I I I I I I I I I I

Interest Rate Percentage . . . . . . . . . . . . .
Average Number of Days in Inventory . . . . . . .

Stacking & Handling Costs for the highest grade of
lumber used S/MBF _ , ,

Gluing cost 5,265. I I I I I I I I_ I I I I I I I

other “at, I I I I I I I I I I I I I I I I I I I

LUMBER AND PROCESSING COSTS AND EFFICIENCY BY GRADE

7

10

11

12

13

14

15

16

17

FAS cos c s IP18? I I I I I I I I I I I I I I I
Cutting 1abor & overhead cost S/HBF . . . . .

Plant yield adjustment percentage . . . . . .
2 plant yield adjustment for lengths over 50"

SEL Cos: SIMBF . . . . . . . .
Cutting labor 5 overhead cost S/MBF . . . . .

Plant yield adjustment percentage . . . . .
I plan: yield adjustment for lengths over 40"

1C Cost SlmFI I I I I I I I I I I I I I I I I
Cutting labor 5 overhead cost SIMBF . . . . .

Plan: yield adjustment percentage . . . . . .
2 plan: yield adjustment for lengths over 60"

2C Cost S/MBF . . . . . . . . .
Cutting labor & overhead cost S/MBF . . . . .

Plan: yield adjustment percentage . . . . . .
1 plant yield adjustment for lengths over 40"

SAC cos: s/mp I I I I I I I I I
Cutting labor 6 overhead cost S/MBF . . . . .

Plan: yield adjustment percentage . . . . . .
2 plant yield adjustment for lengths over 40"

1c and Better slyBF I I I I I I I I I I I I I
Cutting labor & overhead cost S/MBF . . . . .

Plant yield adjustment percentage . . . . . .
2 plan: yield adjustment for lengths over A0"

FIGURE 5.2 BLANK WORKSHEET FOR THE

CUTTING COMPUTER PROGRAM

;; :.{..-...__;
;; or __
: : :né ~~~—-—

12 a..."

. .13 *rI—II—Iu—

I I I -8“

.14{~~~
I I I15{ U-“—
C I :“16{_-~

o o 017{r‘—0~_:.H

I I I18{—~_

OPTIMUM FURNITURE

w

L c L

I o I

N D NUMBER N LENGTH
a 2 PIECES E INCHES
193$ _____ 20 __
2133 _____ 22 _-
2333 _____ 24 __
25%é ___-_ 25 __
27- _____ 28 __
2983 -____ 30 a-
3121 _-___ 32 _-
3323 _____ 3a _-
3522 _____ 36 __
37$£ ____- 38 -H
39$$ _____ 40 -H
Alli __-_H 42 __
43$3 _____ 44 _-
«5&3 _____ 46 __
47%% _____ as “H
49- H____ 50 __
Siij __-__ 52 __
53%; _____ 54 _H
55 H __-_H 56 __
5772 _____ 58 __
5921 H__-_ so __
6127 H__-_ 62 __
63T3 _____ 54 -H
6577 __-__ 66 -H
6777 _____ 68 __
5923 -____ 7o -_
7111 ____- 72 -H
7333 _____ 74 __
753} _____ 76 a“
7743 _-___ 73 __

SALVAGE LUMBER INFORMATION

79 Value in $/‘ {BF (less than lumber cost)

31

1/8"

lllllltlllllllllllllll[[[Illll

WIDTH
INCHES

Ha
H—s
“a
“a
a“
“a
HH
a“
flu
Hid
u—ae—s
Ha

HH

Length in inches (less than cutting bill) 79
Amount actually used in Z . . .
SPECIFICATION OF DESIRED

GRADE USAGE

 

80

81

82

83

* l_- Maximum amount; Z.- Actual amount; 3.- Minimum amount; 4.- Z of Grade

Specification Code * . . - -
Grade ** e e e e e e e e 0
Specification level (BF or Z)
Specification Code * . . . .
Grade ** . . . . .
Specification level (BF or 2)
Specification Code * . . . . .
Grade ** . . . . . . .
Specification level (BF or 2)
Specification Code * . - -
Grade ** . . . . . . . .
Specification level (BF or Z)

I I

80

. 81

I I [ l I

.e 82

2 83

1/8"

ENTER:

0:

[[[lllllllllllllllllllll[[lIlI
[[[[[[l[[[[[[[[[[[[[[[l[[[[[[[

[
[

[

I

HHHHH

“Hid-“

HHE‘HH

HHHHH

** l.- FAS; 2.- SEL; 3.- 16; é_- 2c; a.- 3AC; at- 1c and Btr
0mm OPTIONS

Output in addition to basic cutting summary.

FIGURE

Marginal cost and board feet needed from each board - - - 34

Break-even grade information

5.2

(Continued)

(l.- yes; O.- no)
84 Details on boards cut from each grade . . . . . . . . . .

Random
1 - Specified

I—I

H

1/79

W 0 R K S H E E T

Optimum Furniture

GENERAL AND COMMON COST INFORMATION

State H xc HLQAN

Company ikm Dee Ldooa 'Pacbcxha

l

Species

'Izéfb»<:}Au

Thickness in l/STs . . . . . . . . . . .
Delivery Cost in S/HBF .‘. . . . . . . .

Drying Cost in S/MBF . . . . . . . . . .
Shrinkage in Z . . . . . . . . . . . .

Interest Rate Percentage . . .
Average Number of Days in Inventory

Stacking 8 Handling Costs for the highest grade of
lumber used SIMBF . , ,

Gluing Cost SIHBF

other costs 0 C I C O O O C O C Q

32

LUMBER AND PROCESSING COSTS AND EFFICIENCY BY GRADE

7 PAS Cost Slnar . .

10

ll

12

16

17

18

SEL

lC

3AC

1C and Better

Cutting labor & overhead cost SIMBF

Plant yield adjustment percentage . . . . .
over 40"

1 plant yield adjustment for lengths

Cost SIMBF . . . . . . .
Cutting labor & overhead cost S/MBF

Plant yield adjustment percentage . .

Z plant yield adjustment for lengths

Cost S/MBF . . . . .
Cutting labor & overhead cost $/M3F

Plant yield adjustment percentage

1 plant yield adjustment for lengths

Cost S/MBF . . . . . . . . . . . .
Cutting labor & overhead cost SIMBF

Plant yield adjustment percentage

1 plant yield adjustment for lengths

Cost $/‘EBF . . . . .
Cutting labor & overhead cost SIMBP

Plant yield adjuStment percentage
2 plant yield adjustment for lengths

3,38? I O O O O 0
Cutting labor & overhead cost

Plant yield adjustment percentage
2 plant yield adjustment for lengths

s/sti- '

over 40"

over 40"

 

Cutting Computer Program

 

Date: .
Bill 0: 1
File“ #2 3

FIGURE 5.3 SAMPLE OF COMPLETED WORKSHEET FOR THE
OPTIMUM FURNITURE CUTTING COMPUTER PROGRAM

Enter all data with

01{F§'£_£ .34 a
52

ll
-— underneath.

LA‘...

33

CUTTING 3 I11.

 

L c L
I o I tun-m:
N 0 NUMBER N Lamar}! mom 0 - Random
2 : PIECES E INCHES 1/8" INCHES l/8" 1 - Specified
1931 _-é3§ 20 ffi 2 -3 .1 i
213: -Lége 22 £9 ; _£ .£ L
2323 -Lfies 24 ii i -i 4 L
25 .91: -4929. 26 1.9. 1. -3. 7 J.
2725 __192 28 L2 3 _£ :2 A
29 .93 “2.29. 30 1.9.: a -1 a L
3131 _____ 32 __ _ __ _. _
3323 _-___ 34 _- _ __ _. _
3532 _____ 35 __ _ __ ._ _
37 ._1..2 _.._.._...._. 38 ...._. _. ._.... _. ..
41 2.3. _._.._....._. 42 _.._. ._ ._.._. .. ..
1.3 3.3. ._.._._._._. 44 .__. _. ._._ ._. _.
45$: ___-_ 46 __ _ __ .s _
47J3 _____ as __ _ _- ._ _
49J£ -____ so __ _ __ .. H
53.1.1 _____ 54 __ _ __ _- _
5513 __-__ S6 __ _ __ .. _
5773 _____ 58 __ _ __ .. _
5931 _____ 6O __ _ -_ ._ _
"2'
61- -____ 52 _- _ __ .. -
63~21 _____ 64 __ _ __ _. _
65' 7" __.._.__.._._. 66 __.._. .. ._._. .. _.
673; ____- 68 __ _ __ .. _
69;- _-___ 7O __ _ __ ._ _
71 3.1 _..._.__.._.._. 72 ._.._. _. ._._. _. ..
73$J _____ 7a __ a __ ._ _
75 3.2 ._.._.._._.._. 76 ._._ _. ._.._. _. ..
77 .19 _._.._.._.._. 78 _._. _ ._._. ._. _.

SALVAGE LUHB ER INFORI‘ATION

 

79 Value in S/MBF (less thanximmcc cost). __£E£Z
Length in inches (less than cutting bill) 79 JLSL
Amount actually used in Z . . . . . . . . 49.9..

SPECIFICATION OF DESIRED GRADE USAGE
80 Specification Code * - - - - - - . . .
Grade ** e 0‘ e I e e e e e e e o e e a e

8
{Aw
[to

Specification level (8? or Z) . . . . . . _________,
81 Specification Code * . . . . . . . . . . _‘

Grade ** e e a e e e e e a e e e e e ‘ ' 81 _.

Specification level (BF or Z) - - - - - - a...___._.

82 Specification Code * . . .

Grad. ** e e e e e e e a e e e e e e o ‘ 82 _
Specification level (BF or Z) . . . . . .

83 Specification Code * . . . . . . . . .
Grade ** . . . . . . . . . . . . . . . . 33 __
Specification level (BF or Z) - - - . - .

* 1_- Maximum aneunt; 2.- Actual amount; 3.- Minimum amount; 4.- Z of Grade
** l.- FAS; 2.- SEL: 3.- 1C; é_- 2C; 5.- 3AC; é.- 1c and 3c:
OUTPUT OPTIONS

Output in addition to basic cutting summary. (l_- yes; o.- no)

84 Details on boards cut from each grade . . . . . . . . . . 7
Marginal cost and board feet needed from each board . . . 8 ’

Break-even grade information . 1/79

F\

FIGURE 5.3 (Continued)

34

Some of the inputs listed in Figure 5.1 are self-explanatory,
but others require an explanation. The PDelivery Costf for
the lumber is the charge paid to transport the lumber to the
rough mill. Often this cost is included in lumber price and
would not be entered separately for the computer program.
The "Drying Cost" represents the costs associated
with air and kiln drying of green lumber. If the lumber is
purchased dry, no cost would be entered for drying. The
"Percent Drying Shrinkage" incurred in drying provides a
better estimate of the quantity of lumber to purchase know-
ing that the wood volume after drying will be less. The
"Stacking and Handling Cost" for the highest grade of lumber
used normallycovers the labor expense to stack and sticker
the lumber piles. Stickering is a common practice in stack-
ing lumber wherein each successive course of boards is sep-
arated from the next by stickSe-usually l to 2 inch wide
boards about 8 feet in length. Since stacking costs vary
by grade, the computer program makes these adjustments based

on the raw data for FPL 118 (shown in Table 5.1):

TABLE 5.1 STACKING COSTS' ADJUSTMENT FACTORS FOR THE OPTIMUM
FURNITURE CUTTING COMPUTER PROGRAM

GRADE FACTOR
FAS 1.00
Select 1.10
1 Common 1.18
2 Common 1.25
3 Common 1.30

Source: Harsh, Stephen, Henry Huber, Ed Pepke, and Paula
Johnson. Optimum Furniture Cutting Program, A Tel-
plan Prggram, User's Manual. Department of Agricultural
Economics, Michigan State University, East Lansing,
Michigan, 1977.

 

 

35

These factors will increase the cost of stacking and handling
for lower grades; the adjustment is necessary because lower
grades have relatively smaller, more numerous boards per unit
volume which require increased labor to stack.

The "Interest Rate? is used to evaluate the inventory
which necessitates a value for the "Average Inventory Period".
Thus the program includes the inventory carrying charge in-
curred when a company purchases lumber and keeps it through
various production stages.

The "Gluing Cost" should include the cost of the glue
and overhead along with the labor involved in gluing. The
gluing process is used to join random width boards to specific
panel dimensions. If the company intends to glue some of the
pieces in the cutting bill the cost for gluing should be in-
put. The computer program automatically anticipates more‘
glue joints for lower lumber grades and adjusts the costs
accordingly (See Table 5.2 below). These adjustment factors
were derived from FPL 118 raw data on hard maple cutting
yields.

TABLE 5.2 GLUING COSTS' ADJUSTMENT FACTORS FOR THE OPTIMUM
FURNITURE CUTTING COMPUTER PROGRAM

GRADE ' FACTOR
FAS 1.00
- Select 1.10
1 Common 1.18
2 Common 1.25
3 Common 1.30

Source: Harsh, Stephen, Henry Huber, Ed Pepke, and Paula
Johnson. Optimum Furniture Cutting Program, a Tel-
plan Program, User's Manual. Department of Agricul—
tural Economics, Michigan State University, East
Lansing, Michigan, 1977.

 

 

36

The remaining costs in Figure 5.1 are more specific,
i.e., "Lumber Cost by Grade". The cost of processing the
lumber (with overhead applied) must be input for each grade.
"Plant Efficiency Level by Grade" commonly termed plant yield
adjustment is entered for each grade to customize the yield
table predictions for the company. The exact application of
the plant yield adjustment will be discussed later.

The "Cutting Bill" includes the number and sizes of
pieces to be processed. Usually a company will save the lum-
ber which is too small to meet the current cutting bill re-
quirements but which may be used in some future cutting bill.
The volume of pieces, called "salvage" or "offalf, can be
estimated by the OFCCP after inputing the minimum length
saved, the value of a thousand board feet of that length,
and the percentage that will actually be used. To permit
comparisons between companies in the ten rough mill field
tests of this study, salvage production and value was dis-
regarded.

The costs and information described above comprise
the inputs to the computer program. Another data input space
is available for a program user to insert any other cost
specific to the company being examined. '

Once the basic cost data and cutting bill require-
ments have been input to the program the analysis begins.

The computer program flowchart appears in Figure 5.4.
The output information from the OFCCP is listed in

5.5. A sample printout appears in Figure 5.6. The first

37

 

 

 

 

 

 

 

 

 

 

 

 

 

 
  

 

   
     
 

 

 

 

 

 

 

 

   
 

 

  

 

  
   

 

 

 

 

 

 

 

 

 

 

 

 

    
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

L i 1
Input Calculate Calculate
Stage Objective Trade-off
(See pigir Function for "‘ Matrix for
ure 5.1) 7 First Grade Shorter, sub-
------------ sequent
. . . . Include: Cuttings
gigggication Delivery cost 7
. Adjusted for
(Check input) Inventory time Determine
j Stacking cost Salvage
. Adjust for Volume
Correction Inventory time ‘
Stage I
and Grade
Calculate
Lumber cost Salva e
Adjust for V 1 9%
Inventory time a u
Inputs .
. Gluing cost *
atisfactor Ad'ust for
? J . . Option to
Cutting Bill .
Restrict
Volume (BF) Grade Mix
Processing Cost .
7 Overhead Cost
Error Checking Yield (FPL 118)
Stage Shrinkage
‘ Adjustment .
Determine
Best Grade
Mix to
More Yes Satisfy Ma-
Grades ‘ trix Re-
quirements
Design Calculate Output
Linear . ‘ Matrix. Optimum
rProgramming According to Solution
Matrix Cutting Bill Fig. 5,5
‘ Set up Matrix's
Right Hand Side = Requirements
Number of Pieces Satisfie-
Required
Li
FIGURE 5.4 GENERAL FLOWCHART FOR THE OPTIMUM FURNITURE

CUTTING COMPUTER PROGRAM

38

Net Cost of Cutting Bill

Which Grades to Process

Volume of each grade to purchase
Volume to process (after shrinkage)
Expected yield of parts

Yield of salvage

Cost of materials and processing each grade
Parts to obtain from each grade
Critical cuts

Cost per piece

Volume per piece

Volume in each cutting size

Net volume of cutting bill

Percent of total cutting bill represented in each
size '

Total amount of lumber to be processed
Overall yield

Overall salvage yield

Value of salvage pieces

Breakeven grade and costs

FIGURE 5.5 OUTPUT INFORMATION FROM THE OPTIMUM FURNITURE
CUTTING COMPUTER PROGRAM

39

00¢ OPTIMUM FURNITURE CUTTING COMPUTER PPOEPRH 099

SPECIES 3 RED ORK -- 4/4 THICKNESS

9 COST SUMNRRY 0
NET COST OF CUTTING BILL S

0 GRRDE UTILIEHTIOH SUMMHRY e

2684.

HMOUNT OF EPRDE 1 COM TO BE PUPCHHSED =

HMOUNT TO BE CUT IN MILL = 1403.
PERCENT YIELD FPOM GRADE = 61.3
SRLVHEE YIELD FROM GFHDE a 7.3
COST 8 S 99?.
CODE LENGTH wInTH P/S C 0 NO. CUT
1 43 0r3 2 4x3 3 YES 625.
2 30 2/8 2 4/8 3 411.
3 3' 3/8 a 4/3 S 154.
4 18 7/8 2 4/3 S 114.
5 16 3/8 3 4/3 S 54.
6 13 3/8 a 4/8 3 80.
9999 10 0/3 3 0x3 P 531.

AMOUNT OF seems 2 con TO BE PURCHRSED

HMOUNT TO BE OUT IN MILL 8 37:0.
PERCENT YIELD FFOM GRHDE a 43.3
SHLVHGE YIELD FROM GFHDE 8 12.?
COST 3 $ 1716.
CODE LENGTH MIDTH P/S C C NO. CUT
a 30 3/3 a 4/3 3 YES 1339
3 20 3/3 a 4/8 3 YES 1646.
4 13 ?/8 a 4/3 8 YES 336.
5 16 3x8 2 4/8 2 YES 346.
6 13 3x3 2 4/8 E 130.
9999 10 0/3 3 0r3 R 3255.
9 DOS HND BORED FEET INFORMATION 0
CODE COST/PC BF/FC TOTRL BF z BF
1 1.09 0.33 521. 15.?
a 0.48 0.53 945. ' 3:30.?-
3 0.39 0.41 ?30. 22.4
4 0.32 0.33 388. 10.0
5 0.23 0.33 114. 3.5
6 0.01 0.33 46. 1.4
9999 0.01 0.21 533. 1?.8
TOTRLS 326?. 100.0

0 TOTRL YIELD HND SHLVHGE INFOPMHTION e
TOTHL RMOUNT TO BE CUT = 5133.

PERCENT YIELD FORM RLL GPRDES

8
SHLVREE YIELD FPO" RLL GPHDEE 8

VRLUE OF SRLVRGE PIECES = S

9 BREfiK-EVEN GPRDE INFORMHTIOH e

GPHDE BE COST
FEE 596.39
SELECT 531.63
3H COM 78.3?

1514. BF

4000.

BF

FIGURE 5.6 OPTIMUM FURNITURE CUTTING COMPUTER

PROGRAM SAMPLE PRINTOUT

40
value output is the "Net Cost of the Cutting Bill? which
includes lumber, processing, and overhead costs of pro-
ducing the cuttings. This cost is termed fnet" since it
is the total cost reduced by any salvage value. Again, in
this study, salvage was not included in any of the costs for
the ten rough mill studies.

Next, the program output shows the grades to cut ac—
cording to the various conditions involved in the model. For
each grade, the volume to purchase is shown along with the
volume to cut, i.e., the purchase amount reduced by shrink-
age. The grade's yield is shown both for primary parts and
for salvage. The cost of lumber, processing, and overhead
are combined into a single cost for each grade.

After displaying the cost and yield information for
each grade, the computer lists the length, width, and num—
ber of pieces it processed during simulated cutting of the
lumber. The widths are coded "R" for random widths to be
produced by gluing boards into panels or TS" for solid widths.
If a "YES" appears under the "CC", critical cut heading, the
computer had to obtain lumber to produce some or all of the
pieces of that size. A manufacturer would be aware that the
critical cuts are the more difficult pieces to obtain.

Following "Cost and Board Feet Information" heading,
the marginal "Cost per Piece" is listed. The marginal cost
would be the incremental cost incurred to produce one ad-
ditional unit. For example, if the cutting bill called for

625 pieces of part number one, the cost of material, labor,

41
and overhead for the 626th piece would be the marginal
cost.

The size of each piece and its relative importance
(expressed in percent of the total bill) is listed. Total-
ing the volume per part gives the net board foot volume
produced. The computer program also totals the amount of
lumber to cut from each grade into a single overall lumber
consumption value. The yields for the individual grades are
combined by a weighted average into a single overall yield.
The salvage yields per grade are similarily averaged. The
salvage yield from all grades is computed along with the sal-
vage value. The salvage value is used to reduce the total
cutting bill's cost to give the fnet cost of cutting bill"
discussed earlier.

Last, the available grades not selected by the com-
puter in the current analysis are listed along with the
"Break-Even Costs." The break-even cost is the price at
which the company would have to purchase the lumber to have
that grade become one of the program's chosen grades.

This section has described the inputs, function and
outputs of the OFCCP. The program is basically a tool to
permit analysis of rough mill operations. The interpretation
of the results must be made by the user to maximize the pro-
gram's value. The use of the Optimum Furniture Cutting Com-
puter Program for analyzing ten rough mill field studies will

be discussed below.

CHAPTER VI

PROCEDURE

In order to accomplish the forestated objectives,
ten rough mill field studies were performed. The Optimum
Furniture Cutting Computer Program (OFCCP) described in
Chapter III was used to evaluate the information gathered
in ten yield and cost studies. Before enlisting the aid
of the OFCCP its yield predictions had to be compared and
validated with the Charts for Calculating Yields of Hard

Maple Lumber, FPL 118.3

Manual Verificgtion of the
Optimum Furniture Cutting
Computer Program

 

The basis for the OFCCP is the USDA Forest Products
Laboratory yield tables (nomograms) published in FPL 118.
The nomograms were computerized and then made accessable
through the Optimum Furniture Cutting Computer Program. To
check both the computerization of the yield tables and the
total interface between program user and the computer pro-
gram's data bank, manual calculations were performed. Com-
parisons were made between manual use of the nomograms and

the computer's results from the same nomograms.

42

43

Manual verification began with the simplest cutting
bill possible: one piece of given length at the standard
width (2 inches, for other widths an adjustment is necessary
from the standard). Originally, only four hard maple grades,
FAS, select, 1 common and 2 common were checked; further
program modifications entailed checking 3A common, black wal-
nut grades, and alder grades. Black walnut and alder were
added to the program because the National Hardwood Lumber
Association inspection rules are significantly different for
these species. Most other commercially used hardwood lumber
species follow the same rules as hard maple.

The results of manual verification of the OFCCP stem
from calculations of single piece cutting bills and multiple
piece cutting bills. Using FPL 118, the yields for length
and width combinations could be calculated and compared to
computer predictions.

The results of comparisons on one piece cutting bills
between the FPL ll8 publication and the computer's pre-
diction were satisfactory; the computer prediction fell with-
in one percent of the manually calculated prediction. This
was true whether computer predictions were compared to the
charts in FPL 118 or to larger scale blueprints.' Regardless
of whether the five percent "rule of thumb" or the four per-
cent "money clause" discussed in Chapter VI is used, the
difference between computer predictions and FPL 118 pre-

dictions is within grading rule tolerance; the difference

44
in yield percentages could also be due to rounding and
transcribing error.

When more than one piece of a particular size is
in a cutting bill, the FPL 118 publication is not designed
to determine yields (unless the additional quantities are
proportionally equal, i.e., 50 pieces of each size or 10,000
pieces of each size, etc.). The linear program portion of
the OFCCP becomes important when various quantities of multi-
ple cutting bill sizes are required. Linear programming uses
the various quantity requirements as constraints to work
around, to achieve an optimal solution. However, it was
always possible to arrange the quantities required of various
sizes in the cutting bill so that computer program predictions
would exactly equal FPL ll8 predictions.

Based on the comparisons between computer yield pre-
dictions described above the author gained confidence in the
abilities of the OFCCP. Addition of black walnut yield charts
and red alder yield tables dictated additional manual veri-
fication by the same methods used for hard maple (FPL 118).
Both species' computer predictions compared satisfactorily
with the respective species' yield tables and width adjust-
ment tables.

Field Testing of the Optimum
Furniture Cutting Computer Program

 

Once the accuracy of the computerized nomograms had

been established, the OFCCP was demonstrated to several

1+5
woodworking companies. As a result of these manufacturers'
suggestions, minor revisions and additions were made to the
output format to allow easier interpretation. The stage was
now set for formal field tests at ten companies requesting,
i.e., volunteering for, rough mill studies.

A field test followed a standardized procedure to
investigate the processing of lumber in a rough mill (see
Figure 6.1, "Flowchart of a Rough Mill Field Study"). Each
study or test was targeted for 20,000 board feet equally
distributed among the separate grades. Some companies pro-
duction rates were so low that two weeks or more would have
been necessary to completely process 20,000 board feet. While
a three day limit was placed on each study, it was sometimes
exceeded by a day to complete a study.

The company was requested to have a National Hard-
wood Lumber Association certified inspector grade and meas-
ure the volume (scale) of the incoming lumber. To eliminate
potential drying degrade, it would have been preferable to
have all the lumber graded when dry but, this was not always
possible. All lumber was scaled dry to eliminate estimates
of volume shrinkage incurred during drying operations. Lum-
ber was segregated into distinct packages by grade (with a
few exceptions where FAS and select were combined). While
it would have been preferable to have each normal hardwood
lumber grade (FAS, select, 1 common, 2 common, and 3A common)
equally represented at each mill, it was impossible to in-

terrupt the corporations' production operations. Some

 

 

Input
Study
Lumber

 

 

1

 

 

 

Inspector
Grades and
Measures
Lumber,

 

 

I

 

 

 

Separate
Grades

 

 

 

 

 

Begin
Field
Study

 

 

 

I

 

Group
Meeting to
Explain
Study

 

 

 

‘

 

 

Begin
Processing
Lumber of

 

lst Station

 

 

 

 

Record
Output

 

 

 

 

Monitor
Gross Time
& Lost Time

 

i

 

 

Calculate
Yield

 

 

[

I
.’

46

 

 

 

_1
Calculate

Production
Rates

 

   

 

 

  
 

 

Station
‘ ?

 

 

Calculate
Total '
Yield

 

 

 

Calculate
Overall
Production
Cost

f

Prepare Data
Summary Table
l "Lumber
Yield Summary"

 

 

 

 

 

 

 

1
Prepare Data

Summary Table
2 "Production
Time Summary"

 

 

 

 

 

 

1
Prepare Data
Summary Table

4 "Comparison
of Computer
Predictions

to Study Re-
sults, PotentiaL
Savings by
Using Computer-
Predicted
Least-Cost
[Grade Choice

 

 

 

 

 

 

Present
Report
to Company

 

 

 

 

Prepare Data
Summary Table
3 "Comparison
of Actual Study
Results to Com—

 

puter Predictions"

 

 

L

 

FIGURE 6.1 FLOWCHART OF A ROUGH MILL FIELD STUDY

47

companies had no need for the extreme high or low grades,
therefore only 1 and 2 common were available at these mills.

Individual groups of graded and scaled lumber were
processed normally through each rough mill. The company's
typical grade mix was processed except when a company desired
to know their production efficiency and production rates for
alternative lumber grades. Because of production demands,
some companies ran more than one thickness during the study;
it was necessary to analyze each thickness separately during
the field study and by computer simulation.

Once the lumber was prepared, processing began by the
company's typical production sequence. Most companies first
crosscut the random length and width lumber to their cutting
bill requirements though this was not always the first cutting
operation. The saw for cutting lumber to length was various-
ly called a crosscut saw, X~C saw, or chop saw. At the first
Operation and at successive operations the usable primary
product, i.e., the "output," was measured by tape measure to
the nearest l/8th inch; the computer program's finest incre-
ment is also l/8th inch. In crosscut-first operations, where
wood pieces are stacked on a skid, hand truck, or pallet, the
volume was bulk measured because the width of individual pieces
varied greatly. Gross volume was calculated via:

Formula:

Length x Width x Height

Gross Volume in Board Feet = of the Stack in inches

144 cubic inches per
board feet

48
Example:

q = 54" x 35" x 60" (" = inches)
78"5 board feet 144 cubic inches
To compensate for air space in loosely piled stacks, the
air space volume was calculated as follows:

Formula:

Length of cuttings X
Cummulative width of air space X
Air space volume=Nomina1 lumber thickness in inches
in board feet 144 Cubic inches per board foot

 

Example:

54" X 127" X l"

47.6 board feet Of air space = 144 cubic inches

Gross volume minus air space volume gave the actual output
lumber volume from the crosscut operation:
Example:

739.9 board feet

actual volume cut = 787.5 board feet ‘ 47-6 board feet

Once the volume of lumber into the saw and the output volume
were known_the yield could be calculated:
Formula:

Outputgproduct volume in board feet

Incoming lumber volume in board feetx lOO

yield in % =

49

Example:

239.9 board feet

1138.3 board feetx 100

65.0% yield =

Crosscutting yield data and the production rate infor—
mation were gathered simultaneously according to standard
industrial engineering practices.33’34’35 The start of
gross (overall) time was defined as the instant the saw-
blade made the first cut. This point in time was chosen
because it was both audible and visible to the researcher.
The stopping time was the instant of the first saw cut in
the next grade or the end of the work shift, whichever oc-
curred first. Both the start time and the stOp time were
defined to preclude judgement by the data collector. Gross
time or overall time was calculated by the following:

Formula:

Gross time in hours = StOp time in hours - Start Time
in Hours

Example:

2.23 hours (total or gross) = 3.23 hours (stopped) -
1.00 hours (started)

The actual production rate was calculated based upon inbound
lumber volume for the first Operation as normally done in

the industry:

50
Formula:
Actual Production Rate

board feet/hour er saw =
@ssuming one saw

Input in board feet
Gross time in hours

Example:

1138.3 board feet

510.5 board feet/hour/saw = 2_23 hours

Production rates are stated "per saw" to enable comparison
between the study companies having various numbers of saws.

Since every plant had a different amount of non-
productive time (down time, or lost time) the idle machine
time was monitored closely by stOpwatch. Down time was re-
corded for any paid period of time greater than one minute
during which the operator was not sawing. Corporate man-
agers and floor supervisors were concerned about the quantity
of non-production time. After determining the total down
time and the causes, the net time was obtained as follows:

Formula:

Net time in hours = non—production time in hours -

gross time in hours.

Example:
2.08 hours net time = 2.23 hours gross time -

0.15 hours down time

The net time and the lumber footage incoming to the
machine and these calculations give the potential production

rate:

51

Formula:

Potential production rate _ Lumber input in board feet/

in board feet/hour/saw " Net time in hours/Number of

crosscut saws

Example:

547.26 board feet =llj8.3 board feet/2.08 hours/l saw

per hour per saw

This rate is termed "potentialf since it would only
be obtained if a company had zero down time. Every company
has a certain amount of paid down time when maintenance is
performed, rest breaks are taken, jobs are changed, and set-
ups are altered. But these mandatory non-productive times
vary from firm to firm and must be considered to calculate
comparable production rates.

Similar data was collected on the ripsaWs; a ripsaw
saws a board to the desired width by ripping lengthwise along
the grain. If ripsawing was the second operation, the lum-
ber volume into the ripsaws usually equalled the crosscut
saw's product. The product was counted by the piece since
each piece had the same dimensions. Differentiation between
lumber grades was possible because after crosscutting the
ends of the boards were spray painted/with a distinctive
color for each grade. In the case of a rip-first operation
or a continuous production line, the line was purged of all
of the previous grade before a new grade was processed.
Yields were calculated for each grade, cumulatively for all

grades, and combined to produce an overall yield of both

52
cross-cutting and ripping. Production rates were timed by

the same method as the crosscut saw rates.

Analysis of the Data Generated
From a Rough Mill Field Study

Most companies easily computed their processing cost
per hour in the rough mill by adding hourly wage rates for
each rough mill employee and direct supervision cost per
hour; also added into rough mill costs were the overhead
costs of power, maintenance, repair, worker benefits (if
not included in wages), depreciation of rough mill equip-
ment, supplies, insurance, and any other rough mill eXpenses.
Factory burden was not be include selling and administration
costs. Once a total cost per hour was known, the cost of

processing each grade was determined as follows:

 

Formula:
Number of hours to
1000 Board Feet
process 1000 board = Production volume
feet (MBF) output per hour
Cost of processing = Number of hours x Cost per hour
].MBF to process 1 MBF for labor and
overhead
Example:

1000 Board feet

1.8 hours/MBF = 547.26 Board feet/hour

$72 per MBF = 1.8 hours/MBF ($40/hour for labor
and overhead)

53

In this example, $72 represents the costs of crosscutting,
ripping, stacking, sorting, and overhead for one thousand
board feet of a specific lumber grade in a hypothetical
rough mill. The costs of rough mill processing were used
in computer analyses.

All of the ten companies used in this study incor-
porated some form of salvage operation to retrieve usable
wood parts after an initial pass through the crosscut and
ripsaws. Not all of the salvage operations were performed
during the study due to time limitations or a particular
company's production sequence. However, when possible the
salvage operation was monitored. In the case of a crosscut
first operation, the spray painted board ends enabled de-
termination of the volume of salvage recovered by grade. Sal-
vage operation yield could not be calculated since the exact
quantity of wood entering the station was undeterminable.
The increased production (due to salvage production) by
grade was easily measured; this figure was then used to cal-
culate overall yield.

Because complete salvage Operations in each study
could not be monitored, the effect of salvage could not be
evaluated in this dissertation even though some data are
included. Thus, the salvage saw was not timed, salvage pro-
duction rates were not computed, and costs of the salvage
operation were not determined.

After each in-plant study, data was prepared and

summarized in four tables for individual company reports.

54

. These tables, which may be found in the appendix by company
number became the raw data for Chapter VII "Results". "Data
Summary Table l Lumber Yield Summary" displays data for each
saw type by grade of lumber. "Data Summary Table 2 Production
Time Summary" is also divided into machining operations and
displays the time data and production rates for each grade.
The OFCCP was used to prepare fiData Summary Table 3, Com-
parison of Actual Study Results to Computer Predictions".

The last of each company's raw data tables is "Data Sum-

mary Table 4, Comparison of Computer Predictions to Study
Results, Potential Savings by Using Computer-Predicted Least-

Cost Grade Choice".

CHAPTER VII

RESULTS AND ANALYSIS

The result of ten cost and yield studies where pro—
duction time and lumber input and output measurements were
taken are summarized below in tabular form. The purpose
of these tables is to support the forestated objectives:
to decide whether the Optimum Furniture Cutting Computer
Program (OFCCP) can produce cost and material savings when
used to predict lumber yields and to aid in lumber procure-
ment. Every company studied had their own particular prod—
uct mix which dictated a cutting bill. The costs of pro-
cessing and materials were obtained from each company's
cost accountants and from the time studies. The cutting
bill requirements and associated costs were used as input
information to the OFCCP.

Analysis of the Optimum Furniture

Cutting Computer Program's
Yield Predictions

 

One of the major values of the OFCCP could come from
its ability to simulate and predict lumber yields. The
yield for each company (output volume divided by input

volume) was tabulated for each grade processed (See Table 7.1).

55

56

an a o
o.no o.~n
n.no o.o
~.na u.«
a.~m n.:~
0.00 o.nn
o.oo o.-
a.~n o.c
w.no n.~
«.30 o.o
~.ac o.o
n.~o s.o
~.:¢ m.a
~.5o n.o
u~.«H vuoam
~_< .>.am
-uu>o gases

N.«o
n.oo

o.«m
w.~s
N.OO
a.m¢
o.~o
n.n0
o.«o
:.na

a.no
n.nm
«.00
v«o«»
«such

as:
o.«

a.“ 3.. $8
<\z.
~.~ o.oh c.a:
<\z
<\z
(\z
¢\z
o.n «.Nn
o.o ~--
<\z
<\z
<\=
a.o «.ma a.oc
ouoam one.“ u..«n

.>«nm lam 0.x

. a
H womofllmwm<fi l I

0.05
o.on

voolfihd: how Coo-an Goadda cacao-n :« mh0«nhha« vOuosvcoo lo«v:u. v

¢.o~ a.on «.55
<\z
a.~ o.oo «.As
“.mn <\: H.o¢
<\z o.oa n.no
n.o~ ~.na n.ca
o.o m.an n.5o
u.“ c.~m o.~a
0.9 ~-- 9.50
sx:
«\z o.oo n.~o
~.~ c.5a m.sh
a.“ o.oo o.~a
n _u u
anon» cu.“ uaodu
.zam yam 0.x
nakum‘u name: I n I
coasoo N

o~paa««an< 00: a <\z ram ouc>«¢m - >A<m taun«¢ - a«¢

choc: Lonla«

«o«u «««l snack COF

.ocuaon

na- usea—3.5 .. 0.x

o«a«uaoauq c«o«» ocean-u. uo coda-unao- .oca« nsoasuocoUN
.ovauu ooo«oc aoe:~o:~«

a.ae
n.~«

0.0.
“.09
~.oo~
o.~o
n.oo~
c.nm

~

e.cn

o.so~
a.sa
u
0.x

e.n~ c.5o
o.o 0.5a
s\z
o.a~ <\z
n.3n “.5v
u.o« w.aa
~.o o.cs
n.o« «.on
o.o ~--
o.o o.oo
<\z n.nc
“.0 m.on
o.o o.so
n n
oao.s odour gags»
.>«dm lam
H.~v~ I
:asao

sf:

~.oo
«\z
<\z

o.na

«.oa
«\z

0.0« «.no

a.o~ <\z
n.a~ n.~“

Cacao m<h :« vac:«oc«
a.so«

«.0 o.oo
«\z
cease n<~ :« savanna”
¢\z a.ns o.no
gouge ace :4 aoussocn
..n~ “.5“ n.~o~
ado.“ u~oaw ago.“
.>~am 3am 0.x

lukcm<l uawm
woo_om

..c~ a.qa o.~o
<\z
<\z
a.~n «\z n.on
«\z
«\z
a... .3“. 1.x
~.s “.50 o.ao~
. «\z
o.o ~o.o~ «n-
«\z ~.¢s c.no
Ao... ~98 ~98
a.« «.5» ~.~o~
u u u

o«.«» 9«o«» v«o«»

.>«sm Ram 01”

PE. 1 -5- u - -
m<h

ondho><

a“:
0.x

01x.

0.x
u:x
0.x
0.x
3.:
0.x
3.:
uax
aux
«nuns
0-x

.«O
as:

“@500

NM.

hens
ncou

ua<¢c rm magma» suntan n.»z<mxou =o<u mo n¢<lxam «.6 unn<h

57

Because different processing sequences directly affect each
station's yield, the company's production method is indi-
cated as a rip-first (RIP) or crosscut saw-first (X-C). For
each grade the yield in the table, is presented for the cross-
cut saw(s), ripsaw(s), and salvage saw.

Table 7.2, ”Summary of Actual Lumber Yields,§ lists
the cumulative yield after crosscutting, ripping, and sal-
vage operation by company and by grade. Each company was
different and produced its own particular products. Seven
companies crosscut-first, one company ripped-first, and two
companies had separate lines for crosscut or rip-first.

Normally furniture and dimension stock producers
procure their grade mix to complement their product spe-
cifications. Once the lumber is in inventory, it must be
matched to the cutting bill. Failure to coordinate cutting
bill requirements with the most economical lumber grade re-
duces yield and increases cost. The most dramatic illustra-
tion of this situation occurred in Company Number Eight's 2
common processing. Because the production of long lengths
(80%, 58% and 52% inches) was low the rough mill foreman
instructed the crosscut saw Operator to cut more long lengths.
The crosscut saw Operator began cutting the long lengths re-
gardless of defects; the crosscut saw yield was 94.3 percent.
However, due to the lack of defect-free rippings, the rip-
saws only obtained a 36.0 percent yield. Together the yield
from 2 common was 29.3 percent. While this is an extreme

example of lack of coordination between cutting bill

S8

amm«:mmm« .mmampm cmawca showman :« muons sons:« coochw: how mowcspm OHOMM «««s zmsou cos nmoxsom
_ canaooopg msz«o> >9 Ommum>< omuzwfimz u .w>< .693

pmsfig 3mm azommouo pew saunas span Ho coapmc«nsoo u gpom goon cumom u mm 3mm psommouo n vux
Bamawm u QHm :ossoo <m u o<n .COEEOO N n on .COEEOO « n o«

xas poo«om pew m<m no «N

_ho:pOWOp oommmoopg «OOHOm cam m<m

H
¢.s 3.0 a.- m.m H.m c.a cospms>mn stacca»m
s.:o o.~: m.om H.mo ¢.Hs o.mm naom .m>¢.ep3
w.no o.om H.mo n.mm ~.Hm m.om nnm .m><.euz
o.mo a.mm m.sm :.mo :.Hs m.ms 0-x .m><.op3

nmo.¢ma omm.oH Ham.sm :mm.m¢ ma¢.- mam.ma :pom Hmpoe
omo.om nmm.~ amm.od cmm.¢H a¢N.H , mam gum Hmpoe
“Ho.moa Ham.u mum.w~ mmm.¢n ~s~.HN som.ma 0.x Hayes

m.mo mam.m «\z <\z <\z ¢\z same m.no «\z <\z <\z «\z gum
H.n¢ noa.s damn m.mn «Hon s.on <\z <\z, <\z ¢\z <\z <\z 0-x oH

s.~s nmm.m «\z <\z smHH n.5m oaan s.Ho snow ¢.Hs sows o.oo 0-x a
m.~o moo.o «\z «\z Ham o.on can: H.os «was o.mn <\z <\z o-x m
O.ow Hmo.: «\z <\z Oman H.mm moo n.mo <\z <\z <\z .<\z 0-x s
m.am me.oH <\z <\z mmws H.nw mod: m.~s mmaam H:.om «masm H¢.om 0-x o
m.mo www.ma mama n.na seam m.mm mama o.mo ommw m.ms noam m.om 0-x m
s.¢w mem.HH mmmm o.om oomc m.mc omoa m.nm «\z «\z <\z <\z an
m.mc an.~H «\z «\z «\z «\z oooo v.00 new: Ho.ou ommfl Ho.os 0-x :
m.~o OHH.HH <\z <\z mmom m.:m new: :.we saws ~.Ha man n.os aux n
~.¢o HHo.mH «\z <\z mssm m.mm swam m.mo wsmoo Hm.mo memoo Hm.mo 0-x m
~.so noa.¢~ NNOH n.~a Oman a.mm nomm o.ns mama m.Hs osoa w.ms 0-x H
a cams»
Hangm>o mm mm a am x mm s mm x mm * amuse
mepo>< os=«o> oe:«o> c«o«» os:«o> OHOH» Oe:«o> U«o«» os:«o> v«m«» os:«o> c«m«» csx head
ooanmfioz Hayes «n <n um um OH O A season pomaom m<m m<m no mum -Eoo

mmHQDEm DamHm HRH: roaom zme B< Qm>mmmmo mQAMHM mmmzba A<=Bo< mo N¢<Z£=m N.m mqm<e

59

requirements and rough lumber grades, other companies also
suffered from the same problem to various lesser degrees.

The difficulty Of matching cutting bills with grades
of lumber available to be cut resulted in a distortion of
the anticipated patterns of highest yields for highest grades
for the rip-first lines. Based on the National Hardwood Lum-
ber Association's grading rules,27 the yields should have been‘
closer to the following:

TABLE 7.3 NATIONAL HARDWOOD LUMBER ASSOCIATION MINIMUM
YIELDS FOR EACH GRADE ACCORDING TO GRADING RULES

Grade Minimum Yield (fl)
FAS 83.3
Select 66.7
1 Common 66.7
2 Common 50.0
3A Common 33-3

But the actual average yields, when weighted by volume of

lumber processed, of the ten studies were as follows:

60

TABLE 7.4 ACTUAL AVERAGE YIELDS BY VOLUME OF LUMBER PRO-
CESSED FOR TEN STUDY COMPANIES (Abstracted from
Table 7.2).

-Grade Crosscut- Rip-first Combined Production
First Yield Yield Processes Yield
(%) (%) (%)

FAS 75.7 70.5 75.6

Select 71.4 71.2 71.4

1 Common 67.4 59.5 65.1

2 Common 57.8 63.1 59.3

3A Common 39.4 50.9 42.6

Weighted

Average 65.0 63.6 64.7 s = 6.3

In order to reduce potential error in the above yields due
to small sample size, the weighted average by lumber volume
processed data from Table 7.2 was used to extract the above
information. Most companies crosscut lumber first, which
explains why the "Combined Production Processes Yield" is
similar to "Crosscut-first Yield". While the crosscut-first
yields and the combined production processes yields followed
the anticipated pattern of higher yields for higher grades,
the rip-first yields did not follow such a logical pattern.
If more than two or three plants were studied for a grade,
the researcher assumes the normal correlation between descen-
ding grade and decending yield pattern would have been ob—
served for the rip-first production sequence as well.

Table 7.5 shows the OFCCP's yields when the computer

was restricted to the same grades processed during the field

61

.mmpmum cm»«:: campmmm :« when: ponsz« coozcuw: sch mm«c:pm U«o«m «««E smack :69

m.mo
m.:m
H.mo
2.0m
m.mb
o.~m
H.Hm
m.am
o.mm
m.¢o
s.oo
n.0o

R cams»
mwmpm><

:oa.~m«
efis.mm
oma.mm

man.o
mam.m
san.m
mna.m
Hm¢.m
mmm.m
mm:.@
mm: .11
NHm.o«
mmm.m
««m.m«
coo.mH

km

empzmamzm5320>
Hamao>o Hmpoe

pmp«ma3mm psommouo
can summwp :aon mo :O«pm:«neoo n :pom pooh canon u mm 3am 930mmouo u oux summwm

xfle nom«mm new m<m mo «N

~mo.a
Hmm.n
Hmm.m

<\z
mmmm
«\z
«\z
<\z
<\z
«\z
ammo
<\z
«\z
«\z
mmmm

mm

«\z
“.0:
«\z
<\z
<\z
«\z
«\z
m.om
¢\z
«\z
<\z
m.m¢

R

www.mm
¢:H.~H
was.w~

a\z
moon
Nam

«\2
sons

<\z

«\z
ommm

a\z
sown

«\z
mmm:

mm

2.3.; 33» 2.33

o<n

o<n

UN

<\z
m.ao
¢.on
<\z
m.mo
«\z
«\z
m.nm
«\z
~.on
<\z
:.Ho

x
cHoH»
om

oao.m:
amm.~a
NNH.¢m

Name

<\z
aNHH
Non:
NASH
mmm:
mmm:
mamm
oomm
omen
meow
once

mm
Os:«o>
o«

umombom

commmoouq mE:«o> an mmmuo>m cmpzm«msu.m>m .cu:

m.mo

«\z
o.co
o.om
m.~m
m.mo
m.mo
:.aw
a.mo
~.mc
H.on
o.Hm

m
cams»

Nom.H~

ONQH
wow
«\z
«anew
seem
«\z
03mm
nmma
mamoc
msam

mm
Os:«o>

nonpmwow can poo«om can m¢m

<\z
Ha.ms
A.Hs
«\z
Ho.mm
m.om
H:.mo
o.¢m

s
saws»

o H pomamm peedmm

nmm.mH
:mm
ao¢.ma

¢\z
<\z
mama
<\z
<\z
«anew
coam
¢\z
:mmH
:mm
mamoo
awed

mm
oe:«o>

m<m

N.::o«pa«>oo
o.~s spam
~.mu mam
m.as 0-x

spam

mam

0-x

«\z mam

<\z 0.x

«.80 0-x
<\z 0.x

«\z 0.x

Ha.mm 0-x
0.05 0-x

<\z dam

Ho.mn 0-x
~.mu mam

as.mo u-x
.H.ss 0-x

m wanna

saws» 0-x

mcm no mam

mmHQDBm aqum AAHE Icaom zme B< QmmmMOOmm mAAHm OZHBBDO
mom qume QWBOHQmmm =<moomm mmeamzoo OZHBBDO MmDBHzmam EDEHEQO mo >m<22=m m.u mqm<9

n max
A

epaocmpm
.w><.cs3
.m><.eu3
.m><.ep3
Hmaoe
Hmpoe
Hmpoe

“Ob-630

HNMJ?

hcma
Isoo

62
study. The computer predicted the logical pattern of de-
creasing yield as grade is reduced.
TABLE 7.6 OPTIMUM FURNITURE CUTTING COMPUTER PROGRAM PRE-

DICTED YIELDS. AVERAGES OF TEN STUDY COMPANIES
WEIGHTED BY VOLUME PROCESSED (Abstracted from

Table 7.5)
Grade Crosscut- Rip-first Combined Production
First Yield Yield Processes Yield
(%)
FAS 71.5 75.2 71.6
Select 70.6 70.8. 70.6
1 Common 66.2 66.5 66.3
2 Common 62.3 54.4 59.0
3A Common 46.1 36.8 42.2
Combined 65.9 58.4 64.1

As in Table 7.2, the above average yields were weighted by
the quantity of lumber required to produce the cutting bill.
The board footage volumes were taken from the computer out-
put in Table 7.5, rather than from the volume values based
on the actual volumes processed in Table 7.2.

While individual company values differed between
Table 7.2 and Table 7.5, the weighted averages of the two
tables are remarkably close (See Figure 7.1). Considering
the five percent "allowable" error accepted in lumber grading,
the basic yield table predictions were correct.

If the computer program's plant yield adjustment is

used, the actual practices of almost any company processing

63

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

100q
90.
\O
801 If; \0 V
l\ \O
’F-i F4 0
l\ l\ O :—1 M
7OJ _.-t:1 m (o I: H
\O \O M O '3' V
—F’ ' ' \O \O
on 0‘ ‘—'—_
60.. m "’
0‘9
'0
H 50. \o N
G) .
OH N N
>- <1- <1-
40. "——
30q
'U '6 "C “C “C '0
<1) d.) 0 Q) Q) 0)
H H H H H H
H U H U H U H U H U H U
20. :13 -H :3 "-4 :3 "-4 cu ~r-l ca "-4 m "—4
:3 “U :3 “U :3 'U :3 'U 3 'U 3 'U
H G) H d) H d.) H G) H G) H d)
U H U H U In U $4 U in U h
10 < Q: < D- < O- < O. < D- < CL.
fl
0
FAS Select 1 2 3A Overall

Common Common Common

Grade

FIGURE 7.1 COMPARISON OF WEIGHTED AVERAGE ACTUAL LUMBER
YIELDS OBSERVED AT TEN ROUGH MILL FIELD STUDIES
(From Table 7.2) T0 WEIGHTED AVERAGE OPTIMUM
FURNITURE CUTTING COMPUTER PROGRAM PREDICTED
LUMBER YIELDS FOR CUTTING BILLS PROCESSED DUR-
ING THE SAME FIELD STUDIES (From Table 7.5)

64
National Hardwood Lumber Association's standard graded lum-
ber can be simulated to within a few board feet, or to with-
in a one percent yield. Therefore, while the averages dis-
cussed above are based on the unaltered yields from FPL 118,
the individual cutting bills produced by each company from
the various grades could be simulated exactly (within one
percent, or a few board feet).

The correlation between higher yield and higher
grades (select yields equal 1 common yields when producing
clear cuttings on two faces) was anticipated, since processing
the highest grades efficiently requires less skill than try-
ing to obtain desired cuttings from lower grades. At one
extreme, an operator can consistently achieve a high yield
from relatively clear FAS, if end trim waste is minimal.

The other extreme, however, occurred more frequently; this
is indicated by the lower plant yield adjustment for 2 and
3A common (discussed shortly), where operators had increased
difficulty sawing. Some operators feel it is almost impos-
sible to obtain clear cuttings from the lower grades, es-
pecially after being conditioned to using higher grades.
The saw Operator accustomed to processing large, relatively
defect-free boards of FAS and select quality must make a
substantial adjustment tx>efficient1y use.2 and 3 common.
Ideally, an Operator looks for the clear areas of a board
in order to capitalize on available cutting opportunities.
Often, the sawyer reverses his approach when he sees only

the defects; thus, he fails to visualize the potential

65
cuttings within the piece of lumber. Only when the machinery
Operators maintain an appropriate perspective are they able
to procure their cutting bill requirements and the highest
yields from low grade lumber.

The concept above is not as illusive as it might seem
superficially. The hardwood lumber grades are based on the
clear cuttings on a board, not on the quantity or size of
defects. In the education and training of a National Hard-
wood Lumber Association certified lumber inspector, the im-
portance of visualizing the clear areas of a board is stressed,
in contrast to looking for defects. To the untrained eye a
board will look like an accumulation of knots, wane, and other
defects; on the other hand to the lumber inSpector, or ef-
ficient rough mill Operator, the same board will reveal some
clear cutting Opportunities. Thus, the importance of per-
ceiving a board from an advantageous clear cutting point of
view can and must be learned in order to maximize plant ef-
ficiency.

Table 7.7, "Plant Yield Adjustments for Ten Companies
by Grade," shows most companies can strive to increase yields
towards the FPL 118 standards. Since the objectives Of this
research do not include identifying methods of yield improve-
ment, the reader is directed to the Rough Mill Operators
Guide to Better Cutting Practices32 for suggestions on yield
improvement techniques.

The objectives of this research do include analyzing

whether the OFCCP can be used to accurately predict lumber

66

.mmom: HonssH Oceanus:

new mcceem cecha cuceccm 2H mamaummmH ececscccc ucHeseu eHcHH HHHe emcee cue ”momsom

chmcHHcc< ch n <\z

wHH Ham Ho chHHmHHEHH :HmcmH OHan pHOHz OH mac OOHCO>OHQ pcpsmsoo hp COHHOHUOHmm

mHH gem Ho mccHHcpHeHH cpwcca chcu chHs ceHAAc>c op eccchcH m<m Ema Hp cccccSHHcH

mmH
H\z
<\z

mm

MES
cccswm:.c<

OHOHH HcmHm
Cossoo <m

mm mm
Nu on
MI! mm
no so
MI! 00H
MII mm
mCHH
HmHHH DIX mm
QCHH
cmH HcAHmueHm mHH
mmm we
fill ®HH
Hm No
.Msv Msv
HCmEPm5.@< PCQEHMS.U<
UHmHW HCNHm UHGHW PCNHm
:oSEoo N :05800 H

.moomo paw
.moomo can
N
mmpmhm HomHom 62m m<m mOOSHOCHH
<\z «\z OH
2 mm a
ma <\z m
<\z <\z H
HOHH HOHH o
mOH mOH m
Hem Hem a
HQH :m m
am am m
on om H
“RV MRV zcmgsoo
Hsospmz.v< HCOEHm:.c<
UHOHM HcmHm UHOHM
HomHmm HamHm m¢m

ma<mw Mm mmHz<mzoo MDDBm zme mom mezmzembhm< QHmHN Bz<qm 5.5 mqm<e

67
yields. Without plant yield adjustments (actually 100 per-
cent plant yield adjustments cause no change in FPL ll8 pre-
dictions), the computer came within two to eight percent of
the actual company yields for each grade. With the appro-
priate plant yield adjustments, each company's yield could
be simulated to within one percent accuracy. Since the lum-
ber inspection may easily be up to five percent off-grade,
a prediction to within one percent should be satisfactory.
Given information on cutting bill requirements, lumber qual-
ity (grade) used, and associated costs, the OFCCP can ac-
curately predict lumber yields.
Procurement of Lumber With

Assistance Of the Optimum
Furniture Cutting Computer

Program

Since the OFCCP can accurately predict lumber yields
it may facilitate the procurement of rough lumber by a sec-
ondary manufacturing company. A lumber buyer needs to know
both the products' raw material requirement for lumber, us-
ually eXpressed in board feet, and the eXpected processing
waste. To determine lumber requirements for a product, the
buyer works from the product specifications to calculate the
board footage volume per unit. Multiplication by the quan-
tity of units to be produced gives the net footage required.
For example:

12 board feet/unit x 2000 units = 24,000 board feet

Most of the loss in processing lumber occurs in the

rough mill; but, the location of waste varies with different

68
companies and even between products within a company. At
one extreme, a dimension producer may have no waste beyond
the rough mill since there are no subsequent Operations.
Conversely, the furniture manufacturer who performs ex-
tensive machining after the rough mill (boring, trimming,
turning, routing, Shaping, sanding, etc.) will incur more
unavoidable losses due to removal of wood, set-ups and a
certain volume of rejects. The informed lumber buyer has
accurate knowledge of where losses occur and how much is
lost at each succeeding production step. To calculate the
amount of lumber to purchase, the yield is divided into the
net footage. To illustrate the above example will be con-
tinued and the overall weighted average yield, 64.7 percent,
found for the ten companies in the research will be used in
the following example:

_ 24Looo net BF
37:09“ BF to buy " 64.7% yield BF = board feet

If the company has additional losses before rough milling,
normally due to drying shrinkage, the lumber buyer must ad-
just the volume purchased accordingly. An informed lumber
buyer will know the shrinkage volumes for each species of
lumber used. The general averages for volumetric shrinkage,
depending on inbound and outbound moisture contents in the
drying facilities, are published in the Wood Handbook 36

and the Rules for the Measurement and Inspection of Hardwood

27

and Cypress Lumber. If an 8 percent shrinkage was antici-
pated during kiln drying, the example above should be modi-

fied as follows:

69

= 37,094 BF dry lumber needed

40,320 BF green (1 - 0.08 kiln shrinkage7

lumber to
buy

The OFCCP automatically outputs the quantity Of lum-
ber to buy based on the cutting bill sizes and quantities,
the yields eXpected from the various grades processed, and
the amount of kiln shrinkage input to the program. Actually
the volume of lumber to purchase is one of the first lines
of output from the program. The partial output format fol-
lows:

***OPTIMUM FURNITURE CUTTING PROGRAM***

SPECIES = HARD MAPLE -- 4/4 THICKNESS

*COST SUMMARY*

NET COST OF CUTTING BILL $10577

*GRADE UTILIZATION SUMMARY*

AMOUNT OF GRADE 1 COM TO BE PURCHASED = 13249. BF

AMOUNT TO BE CUT IN MILL

12374.BF
71-7

(For a complete sample printout see Figure 5.6)

PERCENT YIELD FROM GRADE

In the example printout above the net footage required to
fill the rough mill's cutting bill requirements is 8872
board feet. The computer program calculates the "Amount of
grade...to be purchased" by the same method as the lumber
buyer.

Starting with the cutting bill's information on the
number of pieces required for each cutting and each cutting's

size the computer calculates the net board footage required

70
to produce the cutting bill. By using the yield anticipated
from the grades (chosen by computer or as constrained by the
computer operatOr), the computer program calculates the
quantity Of lumber "to be cut in millf:

12374 rough board feet = 88322%e;i:$:rd feet

This example is from Company Number One where the
rough mill acquired its lumber from their integrated, company-
owned and run sawmill. To be able to order sufficient green
lumber to satisfy rough mill needs from their sawmill the
company had to make an allowance for Shrinkage during kiln
drying. Estimating their lumber shrank an average 6.6 per-
cent during drying Company Number One increased the volume
ordered accordingly. To continue the above example:

13,249 board feet _ 12,374 board feet to out in mill

to be purchased I (l - 0.066 loss to kiln shrinkage)
Hence, the computer program logic used is the same as the
normal procedure used by the lumberjpurchaser.

The OFCCP can be used by the secondary wood products
manufacturing industry for procurement of lumber on a scien-
tific, factual basis. During the time required to gather
and analyze data for this dissertation, many commercial es-
tablishments have acquired access rights to the computer pro—
gram in order to obtain information on whiCh grades of lumber
and what quantities to purchase. Therefore, the Optimum
Furniture Cutting Computer Program has already been used suc-
cessfully by companies processing hardwood lumber to procure

lumber.

71
Analysis of Grade Selection
One of the first uses of the Optimum Furniture Cut-
ting Computer Program was to select a grade or grade mix-
ture for a hardwood lumber processor's cutting bill require-
ments. The computer program analyzes the cutting bill in

light of these constraints:

 

 

Cutting Bill Requirements Constraints

Length of Parts Yield from available

Width of Parts grades

Number of Parts Quantities of lumber
available

Price of each grade
Processing cost for
each grade
Efficiency of pro-
cessing each grade
(plant yield ad-
justment percentage)
In order to analyze the possibility for savings
through a different grade choice, it is necessary to com-
pare the computer chosen lumber grade(s) to the actual
grade(s) of lumber processed during the study. Companies
traditionally have not segregated lumber into distinct
grade packages, although it is sometimes economically ad-
vantageous. To accurately assess the economics involved
in grade selectiOn the combined grades and an overall cut-
ting bill must be used. By computing savings or loss re-
sulting from combining lumber grades and cutting bills, a
more realistic savings value is obtained; and, the inability

of the computer to simulate low grade cutting bills where

the lengths cut exceeded yield table values is avoided.

72

It is very possible for a hardwood lumber processor
to cut sizes in excess of the computer's data bank of yield
table lengths. If a firm produced rough cuttings from a
grade longer than yield table lengths, a computer simulation
was impossible. The yield table limits are matched in the
table below with the cutting lengths exceeding those limits
produced during the study.

TABLE 7.8 MAXIMUM YIELD TABLE LENGTHS FROM FPL 118 VS.
MAXIMUM LENGTHS ACTUALLY CUT DURING STUDY

Grade Maximum Yield Maximum Length
Table Length Actually Cut During
(inches) the Study (inches)

FAS 96 101

Select 96 101

1 Common 80 96,

2 Common 40 81

3A Common 30 43

Potential savings may be eXpressed both as volume
(thousand board feet or MBF) or as a monetary value (dollars
or $1). The savings may also be expressed as a percentage
of improvement over current production practices. Calcu-
lation of potential savings due to choice of the most eco-
nomical grade(s) used came from each company's "Data Summary
Table 4", appearing by company in the Appendix.

The potential savings were determined by simulating
the actual study results for processing each grade. Once
the efficiency level of the company was determined (the
plant yield adjustment percent), the values were put into
the computer program, enabling it to make a choice of grades

for the company's cutting bill. The relative efficiency

73
level for each grade was determined and is Shown in each
company's "Data Summary Table 3" in the Appendix.

The computer program enables the user to make two
different plant yield adjustments. The purpose of plant
yield adjustments is to customize the computer's data base
of the FPL ll8 yield charts for each grade. In the program,
the first plant yield adjustment equally raises or lowers
all yields over the entire range of cutting sizes. For ex-
ample, a plant yield adjustment of 90 percent for FAS would
lower the normal 100 percent plant yield predictions by 10
percent for all lengths. Conversely a plant yield adjust-
ment of 110 percent for FAS would raise the computer's nor-
mal yield predictions by 10 percent.

The second plant yield adjustment is used to affect
only on cutting bill lengths over 40 inches. This adjust-
ment is used in addition to the primary plant yield adjust-
ment. For example, if the FAS plant yield adjustment was
90 percent as in the example above and the plant yield ad-
justment for longer cutting lengths over 40 inches was 95
percent the result would be a 15 percent total reduction
for lengths over 40 inches. This second plant yield adjust-
ment for each grade provides a means for a company eXper-
iencing lower than normal yields of their longest cutting
bill parts to indicate the exact plant yield adjustment
necessary for the computer to accurately predict lumber
purchase requirements, yield expectations, production costs,

etc.

74

Plant yield adjustments listed in Table 7.7 display
the plant yield adjustment necessary to Simulate (to within
one percent) the yield of each cutting bill actually produced
by a company. Some plant yield adjustments are higher than
standard yield table predictions, i.e., greater than 100 per-
cent. Usually when a company achieved yields higher than
the yield predicted at 100 percent plant yield adjustment,
the company was incorporating a limited amount of defects
into their final product. Since FPL 118 is based on a stand-
ard of non-defective sides, ends, and one face, a company
allowing cutting defects would show a yield higher than nor-
mal predictions.

However, most plant yield adjustments fell below
100 percent, indicating the company was not achieving the
yield predicted by the computer (based on FPL 118 data) for
the particular cutting bill requirements and grade of lumber.
The reasons companies failed to reach the standard yield
varied between and within companies: failure to obtain the
computer-predicted yields did not always indicate a poorly
functioning rough mill. While none of the ten companies
included in this research followed this practice, other firms
place stringest quality control specifications on slope in
grain, pin knots, color of wood, sapwood or heatwood, bird
peck, gum streaks, grain distortion, etc. When restrictions
are more stringent than the grading rule guidelines under
which the graphs in FPL 118 were derived, lower yields re-

sult. Company Number One did make three separations of their

75

final hard maple cuttings: normal grain sapwood, normal
grain heartwood, and birds-eye (a grain distortion re-
sembling many little birds eyes). Since each final prod-
uct was combined into one overall production value, Com-
pany One's product separations did not affect their yields.

Table 7.9 lists the "Potential Savings Through
Optimum Grade Utilization". Column 1 shows each company's
annual lumber usage, Obtained through discussions with the
company's management. The potential percent of volume sav-
ings was derived for each company's overall cutting bill
by comparing the actual volume of lumber used in the study
with the amount of lumber which would satisfy the same cut—
ting bill requirements if the Optimum grade mix suggested
by the computer had been used. To facilitate comparisons,
the various efficiency levels for each company's available
grades were set into the program using the plant yield ad-
justment factors discussed earlier. The potential volume
savings ranged from a low to 2.0 to a high of 28.7 per-
cent. The lower values came from companies which already
used close to the computer's optimum grade selection. Higher
potential savings, of course, were shown for companies which
used different lumber grades than the computer recommended.

Since there was such a wide range of savings en-
countered in the ten rough mill studies, the arithmetic
mean of 15.6 percent and potentially unreliable due to a

standard deviation of 8.1 percent.

76

umanEoo umuHu-aH: can 0-x a guom ocHH umHHm-zamon u :_z

ocHxnmchmooona+.HaH xpsum mo umou\mocmuw .uao mchmOOOHQ+H0nsaH mo umoov-«v n HBV sea—cu
mocahm .uao mchmOOOHa can noaszH mo umoc - uchmOOOH: can Hossa« xpsum we umou u now :ssHou
oonHHocma peasaa\mcaea> .co e occaaecc ccaaaao-aou:H«o maea>em beaac> Heaccaccae HNV ceaaco
ocHH umHHu-usommonouu-x zHco cossoo <m 62m .N.Hm pmuaHzeHm on p0: cHsoo oncz Hosum

:H mosaHo> oszoxO has : cesHoo zHco o H + HOOHmmm zHcO o H + mam .m<m zHco xmo o H\

a m H
ooo.mma m.mm Hem moo.nH moo H.om o.m zoom
o.~H mHNH Hoo oH o.o HaHm
o.~o HooH moo.“ H.o~ munx oH
ooo.no o.n~ aoma oom.o mmH o.m~ m.o o
ooo.oo nm.mm moo: mono.m HNNH mm.~H m.o m
ooo.nmm.m a.oo omen Hmo.o one m.om m.m A
ooo omH o.m mmoH mmno m Noom ~o.~ o.oH o
ooo.onn o.o obm mmo.oH omoH o.o o.HH m
ooo.mmH .HH HomH mmm.n~ moH m.m o.~ cecm
m.HH ooHH a HH a.o aHm
m.e Hm: moH.~H . o.~ 0.x a
ooo.om~ m.Hn oocH omo.o moo n.mm a.H n
ooo.oHa o.~H oomm HHo.oH can ~.nH o.n m
ooo.non n.aH mmoa an.mm mom. o.~H o.n H
SENSE: So :3 9:38 E x 3 $3 mmcfien ans: sceaeeo
Haw m cH>mm mmcH>mm COHHmstHm :Hom mm: zHHmscc< OssHO> COHHme:mcoo
Hmoo «moo am mwcH>mm zcspm cH mwcH>mm HmHH:Opom Hmscc<
stcc< HmHucOpom umoo Haves pmnssq Ho OESHO>

OssHo> HaHHCOHom
Hmv HOV “my Hov Any Amy HHV

mHHHZ zozoz >a=hm ZMH b< ZOHH<NHHHH= ma<¢c zazmhmo :o:oz:& mozm><m A<HH2mHOQ a.u mgm<b

77

To estimate annual lumber savings, the annual lum-
ber usage was multiplied by the potential volume savings
percentage. Because the Size of the companies ranged from
one half million to eleven million board feet per year, an
overall average potential lumber Savings must be treated
with caution. The companies' potential lumber savings were
not found to be correlated statistically with either company
size (annual consumption) or with potential volume savings
percentage (See Table 7.9). The annual lumber savings could
be as high as 1,056,000 board feet, as in Company Five; but,
realistically, savings will fall below that value.

Prediction of potential savings deserves discussion
Since the percentages and the volumes are maximums and may
not be realistic. There are various constraints which pre-
vent some of the companies from obtaining maximum savings:
lumber grade availability, plant design and capacity, and
willingness to alter present purchasing procedures.

Company Three, for example, uses a mixture of FAS,
select, 1 common and 2 common aspen lumber. While this com-
pany realizes now that using all FAS lumber would reduce
their cost of raw material per unit of production and in-
crease their production rates, the company cannot purchase
any more than their present volume of FAS lumber. The root
of the problem is the relatively poor quality of aspen grow-
ing stock leading to equally low quality logs and lumber.

The company must do the next best alternative: orient its

78

purchasing procedures to include the maximum amount of FAS
available.

The constraints of plant capacity are important when
a company cannot take the extra time required to process
lower grade lumber to meet their requirements; it takes more
time to obtain required cuttings from lower grades with lower
yields, more lumber must be processed to meet requirements.
Such a conflict was not observed during the study, but it is
encountered when a company's production demands dictate rapid
processing, when labor shortages exist, or when waste re-
moval is a problem. To complete a rush order, sometimes a
manufacturer must minimize production time and increase ma-
terial costs by processing a less-time-consuming higher grade.
If labor shortages or short-run plant capacity restrictions
exist, use of the higher grades may increase production.
The use of only select and FAS occurs in some lumber proces-
sing establishments where waste disposal is a difficult prob-
lem, such as in New York City.

The seventh column in Table 7.9, YAnnual Cost Savings"
was derived as follows:
A Annual Lumber x Study Cost Savings
nnual _ Consumption (MMBF) by Simulation ($

Savings I Volume of Lumber in Study for Costing (MBF)

Because of the relative importance of the three factors used
above to determine annual savings, some companies had high
potential savings values. Company Seven had up to $2,523,000,

' while the rest had lower values down to $49,000 per year. An

79
overall savings average of $474,500 per year per company
could occur, but due to the limited sample size and due to
the high standard deviation, $733,000, the reader is cau-
tioned that the actual value of cost savings through cor-
rect grade choice for a company may be extremely different
than this mean annual potential savings. It is accurate to
state, however, that procurement and processing of the Opti-
mum grade mix may lead to thousands of dollars of cost sav-‘
ings for most companies.

Column 6 in Table 7.9 shows each company's potential
cost savings (in percent) through optimization of lumber
grades. These values were taken from each company's Data
Summary Table 4.

The values for potential cost savings were determined
as follows:

Potential Cost=

Cost of Computer's iGradelg) ($))x100
Savings (%)

(1‘ Cost of Grades Used ($)

Cost includes lumber purchased,pr0cessing cost in-
cluding labor, and overhead;
Cost of Grades Used was determined through computer
simulation of the study.
Cost of Computer's Grades was run at each company's
efficiency level.
Again a wide range of potential Savings was found from 3.6
to 60.4 percent. The wide range stems from the degree to

which a company approximated the computer's grade choices.

80

The overall arithmetic mean of 27.1 percent has a very large
standard deviation of 19.4 percent. As before a more valid
conclusion is that the opportunity for cost savings exists
through use of the OFCCP's grade selection. Actual savings
are a function of how close a company presently comes to the
computer's grade choice and how close the company can come
in the future.
Analysis of Lumber
Processing Efficiengy

Even prior to publication of FPL 118, progressive
lumber processing plants have maintained control of rough
mill Operations by periodically monitoring yields. Some
companies find monthly checks sufficient; others, such as
a dimension producer involved in a highly volatile and com-
petitive market, watch yield levels daily. Through com-
parisons with past yield levels, a company can observe
changes in yield levels. If decreasing yields are dis-
covered, remedial steps may be taken to correct the situ-
ation.

Past studies, including one mentioned earlier in
Chapter II, have found the yield values predicted by Charts
for Calculating Yields from Hard Maple Lumberp FPL 118, to
be a workable standard which hardwood lumber users can em-
ploy tO compare their processing efficiency levels. These
same FPL 118 yield tables became the OFCCP's data base; the
Check for accuracy by manual verification was described

earlier. Since the computer program output predicts yields

81
for a cutting bill and associated costs, a manufacturer can
use these yield values as a standard equally well.

The yield table standards are based upon cuttings
with one clear face; but, some companies' specifications are
different. If a company requires all surfaces of a cutting
to be clear of defects (knots, bark, grain deviation, etc.),
a lower yield than predicted by the computer may be anti-
cipated. Conversely, if the manufacturer allows defects in-
corporated into the product the yields may exceed the com-
puter predictions.

In the secondary wood products manufacturing indus-
try, wide variability exists among individual companies'
products and quality control restrictions. Some of the ten
study companies had divergent product specification which
might affect their final yield. Therefore, in analyzing
potential improvements in efficiency (yield) level, it is
necessary to analyze some of the individual company results
(See Tables 7.10 and 7.11).

Company Number One used a grade mix which included
every normal hard maple grade: FAS, select, 1 common, 2 com-
mon, and 3A common. The grade mix was "log run,f meaning
the firm processed the entire lumber production from the com-
pany's captive sawmill located adjacent to the rough mill.
When a rough mill buys a log run lumber, it theoretically
receives all the lumber produced from the saw logs, i.e.,
all high and low grade lumber. Company One's rough mill

study was part of a sequence of studies which also involved

82

.mnmm: Hm953H poozcmmc
pom mcemem omeHco cacpmmm cH onoHumnoH ocpcaoccc mcHoaHm oHcHH HHHE swoon ace

"momzom

.zHco cossoo H can HOOHmm .m

.zHcO mocmmm Cossoo H ccm HOOHmm

.m<m .N

.zpspm CH mssHo> HmHOH smzp mmmH on hme mwcH>Mm mcHEHOHmc OH com: mssHo> .H

m.d
N.¢N

N.om
:.Nm
0
o
o.m

O
o.mH

o.mH
m.H
o

2: 3:33 -o

ooo.no
ooo.omm

ooo.omm
ooo.omm
o
o
ooo.oom

o
ooo.umH

ooo.HmH
000.0:
0
SEC

Amy Ammo mmcH>mm

L mwcH>mm
ossHo>
HmHHQOHom

on

mESHo>
Hmscc<
Hmecceca

Amy

mHm>MH DHMHM wHH Ham 0% HozmHOHkmm

Nm¢.o
mm:.o

mum.H
moo.H
wOH.o
mwo.o
Hmo.o

mmH.H
mmH.H

mmn.o
omo.o
som.o

Hogan
:H COHHQesm
1:00 Honesq
Hmscc< Ho R

on

Hmm mam.o

«Ho.H moo.m

mmm.a som.m

oac.m moo.m

o Hmo.o

o mom.o

oom mno.o
o mca.oH

mmo.H NHm.oH
onm.H amn.o

oon HHm.oH

o coo.mH
ANV-AHV Ammo
Ammo spawn oHH Sam
2H nmcH>mm sooa Hm poo
wasHo> OH m83H0>

HmHeccpca

Amv

moo.b
moo.n

wmm.o

momo.w_

Hmo.o

mmmm.m
Nmmm.m

non.HH
moH.mH

oHH.oH
HHo.oH
mbH.aH
an

:H 930
mESHo>

AHV

aHaw

o.x a
m
N
H

acmmsoo

UZHm¢mmozH xUDomme mozH><m MEDHO> H<HBszom OH.m mHm<B

83

o.m
m.nm

m.mm

o.0

o.o
o

0

0
m.w
0.5H
w.H
n.0H

8220\1870
ARV

mmcH>mm

enco

Hmecmeca
Hoo

ooo.om
ooo.oom

ooo.ow
ooo.Hm
ooo.omm
o

o

0
000.05

omm.mmH
ooo.mm
000.00N

also

u>mm Hmoo
Hmscs<
HmHucceca

Hmv

Noo.o
mm:.o

osm.H
moo.H
ooH.o
omo.o
Hoo.o

mmH.H
50H.H

mmn.o
omo.o
aom.o

spawn

CH COHHmssm
I200 Honssq
Hmszc< HO &

on

.mpmm: HOnEsH coozpmm:

ccH magnum occho camemmm cH onoHunnoH cancaoccc ucHoaHm oHcHH HHHE swoon ace
.zHCO Cossoo H + HomHom .3

"Mombom

.tho Coesoo H.+ HomHmm + 04m .m
.zpzpm Ho cOHHmHssHm 00950500 an cmchHno .N
.msz> mmm>HMm Ho: Han .vmmnpm>o +wchmOOOHQ +HOQE:H mmszosH Hmoo .H

00H
00H.H

omH.H
mom
mom

o

o

o
moo
mom
mmm
mmm.H

Hmo-AHv
AHV Hogan
:H meH
I>mm pmoo
HmHHCOHom

Hmv

mma.m
eon.m

mom.m
omo.m
an.o
on.n
aom.o

NHm.oH
HHm.o

mam.o
moo.Hm
on.oH

n; a:
0HH Hmm

ROOH

Hm Hmoo

Hmv

Hoo.m
ono.a

mmo.o
omm.m
mHo.a
onm.o
mom.o

30H.0
ooo.oH

omm.m
omm.Hm
Hao.HH

N.HAHV

Henssq
scram

Ho pnco

AHV

aHm
0.x oH

m

hcmasoo

mHm>mH QHMHM mHH Hmm 08 wozmHonmm 02Hm<mmOZH musomre mqu><m Emoo HdHBzmeom HH.m mqm¢e

84
studies of the firm's sawmilling and lumber drying efficiency.
Because of the method used to measure lumber volume at the
plant, the yields were near or above 100 percent after cross-
cutting: The firm's production sequence relied heavily on
their salvage operation which increased yield by 8.5 percent.
Salvage costs, volumes, and yields were deleted from computer
simulation as explained in Chapter V. Exclusion of salvage
output and erroneously low input volumes caused low proces-
sing yields. As a result, Company Number One's efficiency
level appears relatively high because little lumber was re-
quired to produce their cutting bill requirements compared
to the computer predictions. But, Company One's yield level
was low because their salvage was not included in the rip-
saw product yield. On the one hand, Table 7.10 shows no
lumber savings and Table 7.11 shows no cost savings: how-
ever, the plant yield adjustments (used to shift the yield
tables' predictions higher or lower according to a company's
performance level) superficially indicate that the company

could improve efficiency:

TABLE 7.12 COMPANY ONE'S PLANT YIELD ADJUSTMENTS BY GRADE

Grade Plant Yield Adjustment
FAS 90%
Select 79%
1 Common 92%
2 Common 91%

3A Common 73%

85
Since plant yield adjustments are less than 100 per-
cent, it appears that Company One could improve their proces-
sing. If salvage were included for this company which relies
so heavily on its salvage Operation.¢he compariSon of yields
would Show these results:
TABLE 7.13 COMPARISON OF ACTUAL YIELDS WITH AND WITHOUT

SALVAGE YIELDS INCLUDED TO THE COMPUTER PRE-
DICTED YIELD WITHOUT SALVAGE FOR COMPANY ONE.

Grade Actual Yield Actual Yield Computer Predicted
Without Salvage With Salvage Yield Without
(%) (%) Salvage (%)

FAS 69.3 78.6 77.1

Select 58.4 71.5. 74.0

1 Common 65.6 73.6 71.6

2 Common 56.1 63.1 61.4

3A Common 33.9 42.3 45.3

From the above, it becomes apparent that Company One actually
presently performs at a relatively high level of efficiency
compared to the OFCCP's standards for yield.

Company Two is a dimension manufacturer processing
FAS, select, 1 common and 2 common cherry. The values in
Table 7.10, showing "Potential Volume Savings Through In-
creasing Efficiency to FPL 118 Yield Levelsf, indicate that
the firm has not met the yield table standards. Assuming
the manufacturer could increase yields to meet yield table

predicted values, a 45,000 board foot reduction in lumber

86
consumption would be realized annually. Based on the fol-
lowing calculation, a Savings of 1.5 percent might be achieved:
Volume to Cut at
100% FPL 118 (BF)

Volume Cut in Study
(BF)

% Potential Volume

Savings = (l -

) x 100

Table 7.11, ”Potential Cost Savings Through Increasing
Efficiency to FPL ll8 Yield Levels," places dollar values on
the savings. For Company Two, the difference between the cost
of the study lumber including processing and overhead costs,
and the simulated cost (computed at 100 percent FPL 118) is
the savings potential during the study. If the study savings
are divided by the percentage of annual yield represented in

the study the result is potential annual savings:

Equation:

Cost of Study Lumber & Processing-Cost at 100% of FPL 118
%'of Annual Lumber Consumption in Study

= Potential Annual Savings

Example for Company Two:

fl"§?25;%21’““3 = ...,...

While the $385 potential savings incurred during the
study might look small, the savings multiply to larger amounts
annually. The potential cost savings, calculated in the same
way as percent volume savings, could be 1.8 percent if the

firm increased efficiency up to FPL ll8 yield standards.

87

Company Three had similar circumstances as Company
Two, with the exception of larger potential savings. Com-
pany Four has two production lines: a gang rip first and
a crosscut first line. The crosscut first line produced
wide furniture parts from a 1 common and better grade mix
(FAS and select and 1 common in varying quantities). For
the crosscut first line the computer program's predictions
on yield, associated costs, and lumber quantities were com-
pared with the actual values observed in the study, indi-
cating the firm may reduce costs and lumber consumption if
it increases efficiency to the maximum predictions of FPL
118. The potential savings could be 157,000 board feet an-
nually, or 15.4 percent Of lumber consumption. In monetary
values, these potential savings through increased efficiency
may be as high as $70,000, or a 8.3 percent savings.

While it appears the crosscut-first line could im-
prove efficiency, the ripsaw-first line achieved yields
higher than computer predictions for the same cutting bill
and grades (1, 2 and 3A common). No savings are listed on
either Table 7.10 or Table 7.11 when the company exceeded
yield table predictions.

Company Five came very close to yield table predicted
values, and, in fact, exceeded them in their FAS and select
grade processing. Unfortunately, the FPL 118 yield tables
are not boundless: and, the computer program maintains their

accuracy by not extrapolating data values outside the original

88
yield charts. Company Five out lengths longer than those
covered by tables for 2 and 3A common (their limits are 40
and 30 inches respectively). Thus, calculations to deter-
mine potential savings at Company Five focused only on FAS,
select and 1 common.

The volume cut at Company Five in the study (9975
board feet) less the volume of lumber required at 100 percent
of FPL 118 predictions is 500 board feet. The 500 board feet
translate to 549,000 board feet potential annual savings. No
cost savings were predicted because the high grade yields
and relatively low costs are near the OFCCP predictions.

Company Six performed at a higher level of efficiency
than the computer program predictions. This firm incorpor-
ated a large number of defects in their piano backframes
since they are hidden from view. While no potential savings
were predicted, there could be an Opportunity for increasing
processing efficiency.

Company Seven had a very small sample of 1 common
(683 board feet) and another small sample of 2 common (3336
board feet). While the field study design called for 20,000
board feet Of one Species and of one thickness, this study
was terminated after four day's total of 14,053 board feet.
By the end of the study, two species and three thicknesses
had been processed. In an attempt to reduce the error intro-
duced by different thicknesses and species, the 5/4 oak was

Chosen for inclusion into this analysis; but, the small sample

89
size reduced the statistical validity of the field test.
While no lumber volume savings would be apparently achieved
if the company improved efficiency, the maximum cost savings
possible would be 6.0 percent or $250,000 annually.

Tables 7.10 and 7.11 reveal the opportunity for Com-
panies Eight, Nine,and Ten to achieve both cost and lumber
consumption savings if they could improve their current pro-
duction operations to increase yields closer to computer pre-
dictions. Because of the diversity Of the sample, and the
heterogeneity of the population, an attempt to calculate mean
values from the ten field studies would be statistically un-
sound. The range of savings is reported below:

TABLE 7.14 RANGE OF POTENTIAL SAVINGS THROUGH INCREASED
EFFICIENCY. (Values from Tables 7.10 and 7.11)

Parameter High Low

Potential Annual Volume

Savings (BF) 549,000 0
Potential Volume Savings (%) 37.2 0
Potential Annual Cost

Savings ($) 250,000 0
Potential Cost Savings (%) 23.7 0

The reader is cautioned that the diverse nature of the sec-
ondary wood products industry, coupled with the potential
sources of error discussed previously, limit general ap-
plication of these results. Some companies may exceed the
potential savings listed above, while others with apparent
opportunity to increase efficiency will not obtain the pre-

dicted savings.

CHAPTER VIII

PRACTICAL ROUGH MILL EXPERIENCES
AND OBSERVATIONS

The purpose of this chapter is to document some of
the observations, and some of the practical experiences
gained during the ten rough mill studies. While the dis-
sertation objectives revolved around the Optimum Furniture
Cutting Computer Program (OFCCP), many of the tOpics dis-
cussed within this chapter may not be related to the com—
puter program. Therefore, the ideas covered herein may
be useful for rough mill managers, forest products utili-
zation specialists, and industry consultants, regardless
of whether or not the OFCCP is employed. The chapter has
two main subdivisions: lumber procurement and lumber pro—

cessing.

Lumber Procurement

Procuring lumber is a first step towards manufactur-
ing a final product in a rough mill. Usually the lumber.
buyer has freedom in purchasing lumber, although in some

of the ten study rough mills various limitations prevented

90

91
procurement liberties (captive sawmills, purchase of lum-
ber by parent corporation, and market supply).

The lumber buyer must have the freedom to purchase
the grade, species, and qualtites necessary to satisfy re—
quirements of the products' cutting bills. To minimize cost.
the buyer must maintain as much flexibility as possible in
making purchase decisions. For example, if the species has
not been designated by the designers, or several species
would be equally acceptable, the buyer could minimize lum-
ber expenses by purchasing the lowest cost species.

A similar philosophy applies to the grade or grades
of lumber purchased. Furniture and dimension manufacturers
usually have a choice of five lumber grades to purchase
(FAS, select, 1 common, 2 common, and 3A common). However,
sometimes a premium above the normal lumber price will have
to be paid, if only one lumber grade is purchased (such as
straight FAS): but, it still may be economically advantageous
to pay the premium and process that grade.

The choice of grade is complicated by the varying
yields obtainable from each grade. Probably the greatest
value of a yield study to a rough mill, is determing the
yield percentage from each grade processed. Many variables,
other than grade, affecting product yield must be controlled;
in order to accurately measure yields for specific grades,
the normal conditions should be maintained: processing se-

quence, operators, cutting bills, and species. Perferably

92
more than one yield study should be performed to verify
that initial yields can be duplicated in future processing.
The most cost-conscious rough mill managers perform yield
and cost studies at regular intervals: daily, weekly, or
monthly.

Armed with the knowledge of their company's Obtain-
able yields, the lumber buyer is in a position to Obtain
the best available value. If various constraints are present,
the buyer must orient his order~ as close as possible to the
desired grade mix. Several examples of purchase constraints,
affecting some of the ten field study companies could be help-
ful in understanding lumber procurement.

The OFCCP predicted that Company Three should pur-
chase only FAS lumber. The company uses aspen, which like
some other species such as beech, has a relatively small
price differential between the grades. In the computer pre-
dictions for the least-cost grade mix for the prevailing
conditions, the computer program weighs the anticipated
grades' yields against the relevant costs of lumber and
processing for each grade. Since the high yield of FAS
could be Obtained with a relatively small increase in pur-
chase price, and a reduced processing cost (less labor in-
volved), the computer chose FAS. As mentioned earlier FAS
aspen lumber is not plentiful; Company Three now orients

their grade mix purchases towards FAS.

93

Table 8.1 compares the OFCCP's grade selections,
to the actual grades of lumber used, for the same cutting
bills processed during the ten rough mill studies. It be-
comes apparent from the table, that select grade lumber was
never chosen by the program. Select was never predicted be-
cause the yield of select grade lumber is near the 1 common
grade yield, while the price of select is nearer FAS (usually
select is only about $10 less than the FAS price per thousand
board feet). The select example, and the forementioned as-
pen example, demonstrate the important interaction between
lumber cost and the anticipated yield which is intrinsic to
knowledgeable lumber procurement.

Another valuble aspect of the rough mill field
studies was the identification of the individual grade's
processing costs. Introduction of the OFCCP to a company
has always renewed management's interest in cost accounting.
Often the cost values calculated after a rough mill cost”
study, differed from the manufacturers previously quoted
costs. Precise knowledge of the cost for processing each
grade of lumber is a necessity for intelligent lumber pur—
chasing, even if the OFCCP is not used to assist in purchasing
decisions.

Once the lumber buyer has determined the grade and
species to purchase, the next decision is the quantity of
lumber to purchase. Calculation of the lumber volume to
buy was covered in the preceeding chapter. If the lumber

is green, an additional shrinkage allowance must be included

94

TABLE 8.1 OPTIMUM FURNITURE CUTTING COMPUTER PROGRAM GRADE
CHOIDE COMPARED TO ACTUAL GRADES USED IN TEN
FIELD STUDY ROUGH MILL

Company

lO

lC

Source:

Grade of
Lumber Used

FAS
Select
10

20

3A0

FAS + Select
10
20

FAS
Select
10.

20

FAS and Select
10 (X-C line)
10 (Rip line)
2C

3AC

FAS
Select
1C

2C

BC

FAS + Select
10
20

l and 2 0

Select
10
20

FAS
Select
10

2C

1C
2C
30
= 1 Common

20: 2 Common 3A0 =
Ten rough mill field studies conducted for hardwood

Computer Predicted
Grade(s) at Company
Yield Levels

FAS

FAS + lC
lC + 20
1C + 20
10 + 2C

FAS + 20
FAS + 10
FAS + 20

FAS
FAS
FAS
FAS

FAS and 1C
10 + 20

FAS + 10 + 2C
FAS + 10
10 + 20

FAS + 20

FAS
FAS
FAS

FAS

FAS + 10
10

10

FAS + 10 + 20
10
20
3A Common

lumber using rough mills in Eastern United States,

1977-1979

95

to prevent shortages. Some lumber buyers are forced to
purchase lumber inventory to guarantee sufficient raw ma-
terial stock; others can obtain lumber as needed, usually
when the rough mill is integrated with the company's sawmill.

The cost of carrying high inventory volumes must be
weighed against the possibility of escaping lumber price in-
creases, common in inflationary periods. Company Nine, for.
example, had four year old lumber in dry storage. The com-
pany claimed to have saved money because the price of lumber
had increased greatly in that time; but, the trade-off cost
of tying their money in inventory was not considered.

The purchase of the most economical grade mix, species,
and quantity in these rapidly fluctuating economic times is
a complex procedure. The lumber buyer can no longer be sat-
isfied to simply rely on historic lumber buying practices.
Changes in lumber supply, and changes in lumber cost, will
force use of lower grades, in contrast to the too frequent
consumption of only high grades in the furniture manufacturing

industry.

When lumber is received at a furniture or dimension
manufacturing plant, it normally is accompanied by an in-
voice stating, among other information, the quantity of each
grade shipped. AS each lot of lumber is moved into inven-
tory, maintainance Of accurate records of volume, grade,
species, inventory control number,and location is essential;
with strict inventory control, the lumber can be matched

to the cutting bills for which it was intended later in the

96
rough mill. While most of the ten study firms recorded
inventory volumes and grades for cost accounting purposes,
coordination was occasionally lacking between these op-
erations: lumber buying, lumber drying, and rough mill
processing.

More than one of the study companies discovered
the grades' volumes on the invoice did not match the grades'
volumes once the lumber was reinspected for the rough mill
field study. Therefore, it is important for a secondary
wood products company to periodically, if not constantly,
grade and scale incoming lumber. When descrepancies between
invoice and reinspection are brought to the attention of the
sawmill or lumber broker the differences may be reconciled.
The rough mills which had lumber grade checks regularily
performed, by their company's lumber inspector, rarely were
incorrectly billed for their lumber.

Inspection at the company before the rough mill can
lead to productive communication and coordination with the
sawmill. By reaching an agreement with sawmillers, some
rough mill operations have increased their yields. One com-
pany visited during the same time period the ten rough mill
field studies were performed (but not a study company) had
their yields increased by requesting the sawmill not to edge
the lumber, i.e., allow wane to remain on the edges. A
common sawmill practice to upgrade lumber involves ripping
or trimming a board to remove or group defects: while the

sawmill's product value is increased, the volume (hence

97

potential rough mill yield) is reduced. Some rough mills
outside of this study have been successful in reaching an
agreement with the sawmill to eliminate or reduce extra
trimming, edging, and ripping. Companies have acquired saw-
mills and bolt (less than eight foot long logs) mills to en-
sure a high quality, reduced cost, and stable supply of hard-
wood lumber. Company Four discovered that even an integrated
corporation sawmill operation might not always ship the grade
of lumber as on the invoice.

Coordination with a sawmill or reinspection to the
furniture or dimension company can enable sorting of the in-
dividual grades during inspection. The OFCCP predicts a
grade or grade mix to be processed for a company's product
requirements. Since a considerable savings could result by
processing the predicted grade(s) the company can easily
calculate whether it would be profitable to segregate grades
before the rough mill. It suffices to state that some Of
the largest, most profitable furniture and dimension manu-
facturers in the United States sort their lumber before
processing.-

Procurement of lumber is not a simplistic task as
it is often treated. The conscious lumber purchaser must
coordinate his actions with product designers, industrial
engineer's specifications, kiln operators, inventory-con-
trollers and rough mill supervisors. The job has tradition-

ally been performed without computers,but now computers and

98
computer programs(0FCCP and others) can assist in lumber

procurement and inventory control.

Lumber Processing

Following lumber procurement, drying, and inven-
torying,the rough mill operation begins. While every rough
mill is different,they all perform the similar function of
converting rough, unsurfaced, variable-sized lumber into
semi-finished parts of specified dimensions. Discussion of
lumber processing will center on selection of lumber grades,
production sequences, Operator efficiency, and the salvage
Operation.

If lumber procurement, inventory, and drying are well
coordinated with the rough mill operation, the selection of
lumber to process should be relatively simple. However, the
more common situation involves the rough mill Operators and
supervisors making a decision concerning which inventoried
lumber to process for the current cutting bill. While the
species has been designated in the cutting bill, and extra
lumber may be returned to inventory, the choice of grade to
process can determine whether or not the rough mill will be
profitable. The rough mill must coordinate the grade(s) to
be processed with the cutting bill requirements. The ad-
vantage of using the OFCCP to assist in lumber procurement
was mentioned in Chapter VII; the rough mill may use the
OFCCP to choose the grade and quantity of lumber from in-

ventory too.

99

The sequence of processing in the rough mill is too
often viewed as being determined by present plant layout.
Actually the sequence should be designed to maximize product
yield. Company Four and Company Ten had the flexibility to
begin processing by either crosscutting or ripsawing the
lumber. Company's Four and Ten are dimension manufacturers,
adapting their processing operation to customers' orders.
Although the other study companies included both dimension
and furniture producers, the product had to be adjusted to
the processing sequence; occasionally a higher yield would
have resulted by the opposite production line sequence.

The factors influencing the decision of whether to
rip or crosscut the lumber first include: product dimen-
sions, lumber,and plant eXperience. Products which lend
themselves to rip—first processing typically have many com-
mon, narrow widths such as flooring, moulding, and window
and cabinet parts (stiles, rails, and sashes). For wider
width products a crosscut-first sequence is generally more
advantageous as in furniture casegoods, table, and panel
manufacture.

A flexible rough mill may be able to designate in-
dividual incoming boards for rip or crosscut-first. A
rip-first board might have the majority of defects in a
line, pith running the length of the board, or a long split;
the product yield could be greater by ripping rather than

crosscutting initially.

100

The plant experience is probably the most valuable
factor in deciding whether to crosscut or rip-first. If
a company has been successful in obtaining high product
yields, from the least-cost grade mix of lumber, by their
present production sequence, then a change might only re—
duce yield. However, sometimes changes in products, raw
material lumber quality, species, or a failure to achieve
the maximum yield indicates a change in the order of Oper-
ations should be considered. For more information the
reader's attention is drawn to Henry A. Huber's article
"In the Rough Mill Should You Rip or Crosscut First?"22
Future modifications of the OFCCP will also permit analysis

of whether a particular cutting bill's yield would increase,

and cost would be reduced, by ripping or crosscutting-first.

In the course of the ten rough mill field studies
the researcher observed a wide range of operator efficiency
levels, i.e., production rates, economy of motion, and
accuracy. Table 8.2 illustrates the "Potential and Actual
Production Rates for Crosscut Saws and Ripsaws by Company
and by Grade of Lumber Processed in Ten Rough Mill Field
Studies." The reasons why the production rates differed
between and within companies are too numerous to detail here;
sometimes the reason was not apparent and more studies would

be required to obtain an eXplanation.

101

TABLE 8.2 POTENTIAL AND ACTUAL PRODUCTION RATES FOR CROSSCUT SAWS AND RIPSAWS BY COMPANY AND 3v
GRADE OF LUMBER PROCESSED IN TEN ROUGH MILL FIELD STUDIES .‘ -

Crosscut saws (Based on Input except for Rip-First)

Rip Saws (Based on Output Except for Rip-First)

Over a FAS SEL. 10 20 3A0 Over R FAS SEL. 1c 20 3A0
All% A Rate Rate Rate Rate Rate All 5 A Rate Rate Rate Rate Rate
Co. Down T BHS ans BHS ans ans Down T BHS ans ans ans ans
Time E Time E
1 14.6 P 684.2 1142.3 823.6 553.8 365.3 10 4 P 8 6
.0 , . 3 7. 273.1 388.2 46.0 210.0
x A 642.0 1142.3 618.6 500.6 333.7 A 383.5 258.8 352.7 316.0 150.8
2 26.0 P 411.8l , 342.3 281.4 20.3 P 6.01 264.
x.c A 283.91 “/A 248.9 217.2 N/A 393.11 “/A 210.; 123:; ”/A
3Rip2 14 7 P 253 7 219 1 154 o 161 9 41 1 P A
. . . . . . 994.7 1833.8 1671.4 1 61.4
A 235.8 209.3 132.5 131.8 "/A A 237.7 1619.5 1083.9 1676.2 N/A
1 -
x-c 32.1 P 851.0 7 7.6 .0 2 . 1
A . 581.11 ”/A 5L8.6 N/A N/A 35 A 126.21 ”/A I232; N/A N/A
4 2
RIP 25.4 P 241.1 136.2 99.9 41.6 P 1157.4 44. 1
A "/A "/A 166.0 101.8 76.7 A N/A “/A 372.6 227.8 336.?
5x-c 10.0 P 452.7 496.8 608.6 348.1 293.6 10.7 P 430.0 391.0 9.0 2 0.0 2 .
A 398.1 440.8 553.1 344.1 260.0 A 386.0 355.0 322.0 236.0 233.3
6 _ 16.3 P 1436.31 1342.2 711.6 14.0 P 612. 1 .
x C A 1318.61 ”/A 992.0 603.2 ”/3 A 526.31 ”/A 323.2 363:9 N/A
7x4 19.3 P N/A "/3 39803 82307 N
A . 2 . 6 . ./A 8.4 P 121.1 146.2
95 3 9° 7 A "/A "/A 107.4 135.1 ”/A
8 33.2 P 821.3 708.2 642.1 24 5 P
x-c . 370.4 22 .1 242.
A ”/A 495.6 507.6 338.5 “/A A “/A 225.8 173.8 193.3 “/3
9 2 42.1 P 211.7 198.9 167.1 131.0 --3 P --3 --3 --3 --3
RIP A 97.4 152.4 97;3 71.6 "/A A "/A
x-c 29.9 P 742.3 609.3 20.6 P 122.8
A N/A N/A N/A 562.9 “03 1 A N/A N/A N/A 98.5 133:;
lGRIP2 33.7 P 237 5 33 7 P 983 4
A N/A N/A 157.5 N/A N/A ' A N/A N/A 652:“ N/A N/A
1

Includes select grade

ZOang rip sewn first: rip saw rates based on input lumber volume, X-C saw rates based on output

product volume

3No rip saw rates, automatic line, prevented measurement

P - Potential rate (does not include down time)
X-C I Crosscut-first production sequence.
N/A . Not Applicable

Source: Ten rough mill field studies conducted for Eastern United States hardwood lumber processors,

1977-1979.

581 8 Select 1C I 1 Common

A 8 Actual rate

BHS = Board feet per Hour per Saw

Rip - Ripsaw first production sequence

2C ' 2 Common 3A0 a 3A Common

102

Some general causes of variable production rates
might be helpful to the readers. One of the major causes
of variable production rates is whether the rates were
based on input lumber volumes or output product volumes.
Table 8.2 is simply a listing of production rates; if com-
parisons are to be made between companies, the rip-first
lines in Company Three, Four, Nine and Ten should be anal-
yzed separately from the crosscut-first lines.

Production rates in Table 8.2 are listed as "P" for
potential and "A" for actual; the calculations for these
rates were detailed in Chapter VII. The basic difference
between the two rates is that actual production rates in-
clude the down time (also termed lost time or non-production
time). The potential production rate would only occur if the
company had zero down time. While an unrealistic value, the
potential production rate allows comparisons between com-
panies without the affect of down time.

The "Overall Percent Down Time" listed for both the
crosscut saws and ripsaws gives a good indication why some
companies actual rates are high. The lowest down time re-
corded for saws was 8.4 percent, but a more reasonable tar-
get would be near ten percent. Unnecessary down time was
the major cause of lost production. Some routine maintenance
on the machinery will reduce production, but even this could
be performed by the millwright after hours. TO reduce down

time losses, the Operator should be allowed to run the

103
machine, and not be required to bring lumber to the saw,
take product away, or perform other duties.

Because of time limitations in the study,accurate
measurement of some variables affecting production rates
was impossible.

Variables affecting actual production rates include
the following:

Operator experience and skill

Operator speed

Machine rate

Product size

Grade of lumber

Board size

Length
Thickness
Width

Down time

Sample size

Condition of work environment
Due to the wide range of values Observed for the above list
of variables in the ten rough mill field studies, the reader
should use caution in interpreting and applying the pro-
duction rate results in Table 8.2 For specific suggestions
on improving operator efficiency the reader is again refer-
red to the Rough Mill Operators Guide to Better Cuttipg

'3
Practice s .12

 

104

The final topic to be covered, the salvage oper—
ation, is one which was excluded previously to allow equiv-
lent comparisons between the companies. Table 8.3 "In-
creased Yield From Salvage in Percent by Grade and by Com-
pany from Ten Rough Mill Field Studies? is abstraced from
Table 7.1 (more complete raw data may be found in the Ap-
pendix). Other than when salvage was not processed during
the study, denoted by "N/A" and footnote number two, a wide
range of salvage yields was observed (0.0 to 34.5 percent).

The main reason for divergent salvage yields is the
varying importance placed on the salvage Operation by the
company. At one extreme, Company Four had no salvage oper-
ation. Companies Eight and Nine, on the other hand, placed
a higher priority on the salvage Operation; therefore, the
crosscut and ripsaw Operators allowed more material to be
salvaged into usable parts. Obviously, Company Nine, pro-
cessing high grades for long lengths can achieve higher
added salvage yield than Company Ten, processing low grades
for short lengths (assuming equal salvage lengths for both
companies). Due to the high value of wood; compared to
labor cost, a functional salvage Operation significantly
increases yields for most companies. .

This chapter covered some of the rough mill field
study Observations and experiences gained indirectly while
persuing the objectives of this research. The chapter may

enable understanding some of the variables involved in

105
lumber procurement and processing. Topics covered in this
chapter will not influence Chapters IX and X, "Summary and

Conclusions" and "Recommendations"respectively.

CHAPTER IX

SUMMARY AND CONCLUSIONS

Ten rough mill field studies were conducted in furni-
ture and dimension companies in the eastern United States.
Data was collected to provide the basis for accomplishing
the three study objectives regarding the analysis of the
Optimum Furniture Cutting Computer Program (OFCCP) predictions
for lumber grade utilization and product yields. Because of
the diversity in product, production methods, and raw ma-
terial used by the hardwood lumber processors, the results
in the previous section may or may not be applicable to Sim-
ilar companies.

The first of the three objectives concerns the OFCCP's
capability for predicting lumber yields, given information
on lumber grade requirements, associated costs, and cutting
bill needs. By comparing the computer predicted yields to
the actual yields Observed in the ten studies the computer's
ability to predict yields was analyzed. The OFCCP was found
to predict yields for the cutting bills processed by the ten
companies within a range Of 0.3 to 4.0 percent (on the aver-
age) for the five grades (FAS, select, 1 common, 2 common,
and 3A common). Overall the computer predicted a yield of
64.1 percent which is only 0.6 percent lower than the actual

overall average yield found to be 64.7 percent.

106

107

The OFCCP has a feature enabling the company to
customize the program's yield table data base to the com-
pany's specific yield and lumber volume consumption level.
The plant yield adjustment percentage allows the yield table
values to be increased, decreased, or to remain at the stand-
ard levels as dictated by a company's performance. When the
plant yield adjustment customized the data base for a company,
the yields predicted were always within one percent of the
Observed production values.

Overall, the OFCCP accomplishes the objectives of
accurately predicting lumber yields, given information on
lumber grade requirements, associated costs, and cutting
bill needs. The ability to precisely predict yields, easily
done using appropriate plant yield adjustment percentages,
is essential for procurement of lumber.

Determination of OFCCP's ability to accurately pre-
scribe lumber purchases was the second objective of this
research. The program follows the same procedure used by
lumber buyers in the secondary wood products manufacturing
industry to determine the quantity of lumber to buy. The
program's cost minimization linear program combines the com-
pany's cost, yield, and product requirements data to select
the most economical grade mix for given circumstances. Since
the OFCCP follows normal lumber purchasing procedures and
predicts accurate lumber yields it can and is actually being
used by the secondary wood products manufacturing industry

for the procurement of lumber on a scientific basis.

108

The examination of the Opportunity for financial
and material savings in the secondary wood processing in-
dustry through use of the OFCCP is the final objective of
this research. The fpotentialfi savings were analyzed as a
result of two separate and distinct changes to be made by
a hardwood lumber user. Changed first is the grade mix for
a company's cutting bill with its associated costs. The
second change analyzed was the improvement of lumber proces-
sing efficiency to the maximum level predicted by the pro-
gram.

Because of the diverse production methods, raw ma-
terials, and products of the industry, and of the ten sample
companies, the results obtained coult not be accurately
represented by a mean value. Indeed, the high standard de-
viations incurred suggest that a range of values would best
summarize the data (See Table 9.1).

TABLE 9.1 RANGE OF POTENTIAL SAVINGS INCURRED THROUGH USE
OF THE OPTIMUM FURNITURE CUTTING COMPUTER PROGRAM
AT TEN FIELD STUDY ROUGH MILLS, 1977-1979. (Ab-
stracted from Tables 7.9, 7.10, and 7.11).

Type of Savings Low Value High Value

Optimizing Grade Utilization
Cost Savings

0 306 6004

$ Annually 49,000 2,523,000
Volume Savings

0 2.0 2807

MBF Annually 122 1056

Increasing Production Efficiency

Cost Savings

0 O 23.7

$ Annually 0 250,000

Volume Savings
0 0 37.2
MBF Annually 0 549

109

Table 9.1 displays the wide range of values Obtained
in this study. From this information one can conclude that
the opportunity for monetary and material savings does exist
in some hardwood processing plants. However, specification

to the Size of such savings is impossible at this time.

CHAPTER X

RECOMMENDATIONS

While the Optimum Furniture Cutting Computer Program
(OFCCP) passed its field test demonstration period, the
novelty of the program has now worn off. The advertise-
ment of the computer program has recently been limited due
to potential competition with a similar private program. Most
of the major wood products trade associations have published
news releases and have had presentations on the availability
of the OFCCP. The program's present status for the USDA
Forest Service is that of a tool for use by forest products
utilization specialists when servicing their clientele. No
comprehensive promotion exists to further implement the OFCCP
through the USDA Forest Service, but a growing number of com-
mercial woodworking firms are assessing the computer program
made available by the Michigan State University Teleplan
System.

For the computer program to reach its maximum poten-
tial, a coordinated implementation program would be required.
While the USDA Forest Service is in a position to make the

computer program available to the entire lumber-user industry,

llO

111
the lack of manpower and necessary funds plus the fear of
competition with the private sector prevents implementation
to its fullest degree. Based upon the reception of the com-
puter program by the industry, the need for this tool still
exists. The Opportunity for an enterprising company or in-
dustry consultant to profitably implement the program appears
ripe at the present time. Like many developments in the wood
industry, the economics of the computer program are presently
established; but action may not be taken until the evidence
for potential savings becomes overwhelming. General reluctance
by the secondary processing industry to accept new production
methods or ideas further hinders universal acceptance of the
computer program.

Further research on maximizing utilization needs to
be performed in all of the sequential steps leading to the
lumber processing stage of hardwood conversion. The past,
present, and future research must be of a practical nature
so that results can be successfully implemented by the wood
industry. The various stages which ultimately may have an
impact on hardwood lumber yields are the timber growing, log-
ging, sawmilling, grading, drying, marketing, and lumber
processing. ‘

Tree growing must be performed with the objective
of profitably producing the highest quality timber in the
minimum amount of time. Research and implementation of re-
search on growing and harvesting (to provide information on

maximizing the forest timber resources) must be continued.

112

Recent improvements in logging due to new methods, such as
the USDA Forest Services "Felling and Buckingfi (FAS) pro-
cedures analysis, ultimately boost lumber yields.

The improvements possible in sawmilling have been
initiated through research and implementation of research.
New mechanization has generated increased precision in log
conversion. Improvements in production facilities or
"tightening—up" are being carried out by public and private
industry consultants. Again, increased efficiency in saw-
milling should lead to increased volume production and max-
imum grade production of hardwood lumber.

While grading is fairly universally based on the
National Hardwood Lumber Association's Rules for the Measure-
ment and Inspection Of Hapdwood and Cypress Lumber?7 the
application of the grading rules in sawmilling, marketing
(buying and selling), and processing of lumber should be im-
proved. Processer's needs with lumber grade availability
must be coordinated to maximize lumber utilization.

Research along with implementation of the findings
for lumber drying practices can reduce degrade incurred during
drying. One program used by the USDA Forest Service, State
and Private Forestry is the Improved Drying Program (IDP).
The IDP, as an analysis tool, is used to measure the amount
of drying degrade and then make recommendations to decrease
the losses. The IDP program is a good example of implementa-
tion of research since the user bases recommendations on

findings of earlier research in the lumber drying field.

113

The USDA Forest Service's RIP (Rough-Mill Improve-
ment Program) is another program implementing research on
manufacturing techniques to improve lumber yields, decrease
manufacturing Costs, reduce waste, and extend the hardwood
resource. As the particular stage under analysis moves closer
to the final product, the increases in efficiency become in-
creasingly important. For example, a small percentage in-
crease in rough mill yields will have ramifications (theoret-
ically) back through the successive stages leading to lumber
processing; a small savings in the rough mill could mean
less lumber to dry, less to produce for the company, and an
extension of the hardwood lumber resource. Integration of
the research and research implementation for all the various
stages leading to hardwood lumber processing in the rough

mill would maximize effectiveness.

APPENDIX

FOUR DATA SUMMARY TABLES FOR EACH
PRODUCTION LINE BY COMPANY NUMBER

114

COMPANY 1
RIP COST AND YIELD STUDY DATA SUMMARY TABLE 1
LUMBER YIELD SUMMARY

Cross Cut Saws

 

(1) (2) (3) (4)
Grade Input Output Yield (%)
(BF) (BF) (3):(2) x 100
FAS 1040 1062.3 ' 102.1
Select 1782 1809.1 101.5
1 Common 5765 5642.8 97.9
2 Common' 3956 3670.0 92.8
3A Common 1622 1435.0 88.5
Total 14,165 13,619.2 96.1
Rip Saws
(5) (6)=(3) (7) (8) (9)
Grade Input Output Yield (t) Cummulative
(BF1_ (BF) (2)/(6)x100 Yield (t)
(7)/(2)x100
FAS 1062,3 720,9 67.9 69.3
Select 1809.1 1040.3 57.5 58.4
1 Common 5642.8 3781 0 67.0 65.6
2 Common 3670.0 2218.3 _ 60.4 56.1
3A Common 1435.0 550.3 38.3 33.9

Total 13,619.2 8310.3 61.0 58.7

115

---- m.¢om H.ovm w.oflmw ¢¢.¢N Q.CH mw.N ON.BN Hapop
m.Nm- m.om~ c.oHN m.omm mo.N a.a~ Nc.H mo.m cossou <m
¢.oH- o.on c.84m m.m-~ He.o a.” Hc.o No.5 aoeeoo N
c.o~+ a.Nmm ~.wmm o.Hmam va.a ~.a mm.c Na.ofi :oEEou H
m.~m- m.wm~ H.mam m.o¢o~ Hw.m N.m HN.c Nc.q uuoamm
..... m.mmm 8.5mm a CNN mw.H o.~ mo.o ww.~ m<a
meats flfiasuuav fifiafipcouoav hmgmflmv
Smogwfl: pxoz fizam\gg\amv Azam\t;\amv Amv-fimg mafia Aggy Aggy
zoozuom 1A~Vhflov comm “AmV1AoV comm Ammv meme .czon cane cane
omcmr—U a fioHHUSflOhm :OHHUDGOHQ HSQHSO H02 w ESOD mmOkU owmho
Amm psduso HmuOH com: wommmv mme afim
---- m.com «.mmo moH.VH aw.o~ 0.4H ow.H 40.NH Hapoe
m.mm- “.mmm m.mom NNOH NN.N o.m HN.o m¢.N coesou <m
H.aH- o.oom w.mmm omam mm.m m.cH wm.o oa.m coeeou N
a.mv- o.ch c.mmm moam om.m m.¢~ 0H.H 08.4 :oeeoo H
w.ma+ M.NVHH m.N¢HH Nwafl ma.c o.o oo.o ma.o uumfiom
..... o.Nqo N.¢mo oaoa ca.o ~.o mc.o Hm.o m<a
. mango ”Hasuuav Aflmaucopoav Aggy ”NV1fimV
“manua: uxoz Azam\uc\mmv Azam\u;\umv fimv-fiNv 0509 aged Aggy
:oozpom «AmylfioV comm .nmvafloV comm Ammv mafia czoa meme «sue
omcmzu » newposmopm :oflpustHA usncH uoz a . :309 mmouo owmho
flaw Amy any new nmv flag nmv ”NV HHS

Adm usmcH HmHOH :o wommmv 3mm “:0 mmouu
>m<22=m mZHH onhusaomm
m mAm<H >x<zzam <H<D >Qahm mama» Qz< Hmou mHm

H >z<mzoo

.mcoHumuHEHH cumcoH :oEEoo N m.Emhmopa pousnaoo oprpo>o on wopHscoh coseoo H 6mm moozHocH \M

Q

.mcoHumuHEHH nowcoH :05500 N m.Emhmopm pousnsoo owHupo>o ou wohHscoH :05500 H Nmm mousHocH \M

116

IwmmH o.em lonm mu INcHH m.mv INNNN :oEEou <m m.mm NNcH :oEEou <m
\N \N \N \N .
INmmN m.mm INQNm Hm IHHVN ¢.Ho ImNme :oeeou N H.om cmam :oeeoo N
\H \H \H \H

meme m.mc Hams No comm o.HN ammo :oEEou H o.mo monm :oesou H

HNON m.wm mmom an cmoH o.vm mva uuoHom «.mm NwNH uuoHom

NHHH v.mo «NcH om ooOH H.NN chH m<m m.mc ovoH m<m

Hwy AWL Hwy Ha.
:wauno Hmmv u.He< mHH Sam Hmmv Nzaasoo uocflmpno Hmmv use Remasou
Hwy 0“ “so 08 eHoH> Hay woofi on “so Op Na com: NHHmsou< Nfifiaspu< Na mom:
umoo wHoH> quucmso ucmHm umou wHoH> quucmso mocmho wHoH> popcmso movmuo

 

:oHumHnus waHm

 

 

mcoHuonopm popSQEou mszmom xvsum Hmsuo<

monhonmmm mmFszou Ob mhqammm >QDHm A<DHU< mo zome<m20U
m mHm<F >z<223m <h<a >QDHm quH> Qz< Hmou mHm

H >z<mzou

117

.mcoHumHHEHH camcoH mHnmu wHowx owwhho>o ow COEEOU H wmm wowzHu:H\m
.mcoHumuHEHH camcoH oHan wHon othpo>o ow :oEEoo H mmm wowsHo:H\M

.nopmHsaHm who: mommucoouon nHon cho ooch wow: mosz> ommuoom Hmsuom nouns Ho: oa\w

eHmN 205500 N
wa.NH wN.eH Noe.NH mNH.m~ o>on< HH< Nam.oH eam.NH cossoo H
ommH :oEEou N
mN.NN mm.HN \mmomH onm :oEEoo <m meoH mee :oEEoo H
mHNm :05500 N
mo.o mc.m \MNmmN NeNm :oeaoo N HHeN moNH :05500 H
nmmm coseou N
mN.H wH.NH meme Homu :OEEou H mHmm onwm :oeeou H
HeeH :oaeou H
wH.HN WN.ON HNoN mmom HooHom onH eooH m<m
mc.cH ma.m NHHH eNoH m<m oooH HoeH m<m
oon oon
HHmv\HHV-HH HHev\HNV-H_ HHV Hmmv pcaHg Hwy Hmmv Hmvoemuo
mmcm>mm mm mmcH>mm a Hmou com: :H wow: Hmou wthscom ononopd
HmHucouom HmHucoHom :oHumstHm \HponESH owmho #02534 peasanu
HOV Hmv. Hev Hmv HNV HHS

muHozu mn<mo Hmoo.ew<m4 ampanmmmummHszou osz: >m mozH><m H<thmhom
meqammm >Q=Hm Ob monHuHQmmm zmhzmzoo mo zomHm<mzoo
e mHm<H >m<22=m <P<c >ashm chH> Qz< Hmou mHm

H >z<m20u

118

COMPANY 2
RIP COST AND YIELD STUDY DATA SUMMARY TABLE 1
LUMBER YIELD SUMMARY

Cross Cut Saws

(1) (2) (3) (4)
Grade Input Output Yield (%)
(BF) (BF) (3)4(2) x 100
FAS and Select 12,048 11,281.3 93.6
1 Common 5,087 5,482.2 N/A
2 Common 2,476 1,928.5 77.9
Total 19,611 18,692.0 95.3
Rip Saws
(5) (6)=(3) (7) (8) (9)
Grade Input Output Yield (8) Cummulative
(BF) (BF) (2)/(6)x100 Yield (8)
(7)/(2)x100
FAS+Select 11281,3 7741.9 68,6 64.3
1 Common 5482.2 3008,2 54,9 59.1
2 Common *'1928,5 907,1 47,0 36.6

Total 18,692,0 11,657.2 62.4 59.4

119

N.Hm-
H.wN-

ocmpo

umost: onz.
coozpom
omcmno w

n.NH-
m.NH-

compo

omongz oxoz
:oozpom
omcmnu m

Hay

 

m.wem A.HHm N.Nmo.HH e.am m.c~
o.eeH e.aoH H.5om e.m e.mH
N.¢HN H.eo~ N.mcom e.HH N.cN
H.mm~ o.oam m.HeNA o.o~ o.NN
HHosoooV , HHmHocoooov HNVhHmV
Hzem\o;\mmv Hzmm\o;\mmv HmV-HNV oeHe
«HNVhHoV ooom “HmV1HcV ooax Hmmv oems :30:
_:oHpu:wopm cOHposcomm pzopso uoz »
Hmm poopso coo: eommmv mzmm on
m.eo~ 5.0mm HHo.mH m.NN c.0N
N.NH~ e.HwN caem e.e m.NN
m.we~ m.mem Amcm e.“ m.NN
a.mm~ w.HHe meo.NH o.mH m.c~
HHazoooV HHeHoooooov Huey vauHmv
Hzam\8;\mmv Hzam\oz\mmv HmV-H~U osHe
«HNv1HoV ooom “HmvnHov oomm Hang oeHb czoa
:oHHoswoum :oHuoswogm psch «02 m
Hwy HAS HOV Hmv new

Hum HsmcH Hmwoe com: nomwmv mzmm H30 mmopu
>m<223m mth onHobnomm
N mHm<e >m<zzsm <H<Q

N >z<mzou

>Q3Hm QHmH> az< Hmoo mHm

l\
O)

\OOOM
Lan-i

Hung
oeHe
:Bom

Hmv

szv
oaHe

mmopu

szv

osHe

mmoho

HNV

Hmooe

:05500 N
:05800 H
HooHom
mam m<m

compo

Hmooe

coEEou N
:oEEou H
HooHom
can w<m

ovwhu

HHV

120

ooHnUN
oHHuuH
eauHom
wNw.HN o.mm ooc.mH w w<m mee.HN n.oc HHm.mH o>onm HH<

HHHm onoozu HHooo>o

.mcoHHmHHEHH oHan dHon OH one oHnHmmoa Ho: :oHHmHSEHm :oEEou N
Noe.m H.mm Nwo.m wHH owe.o H.om moo.o coseou H

eom.mH m.eo oom.NH em mmo.eH e.mo NNH.NH oooHom

ecu men
HNV HNL HNV

:Hmono Hmmv H.He< mHH Ham Hmmv HemoEou

HHS oo one oo eHoH> HHS NooH om one oo No eoma

Hmou cHoH> zuHucmso pcmHm umou wHoH> zqucmso mowmho

 

:oHpmHseHm zVSHm

 

mcoHuoHoon Housmaoo

e.mm HHc.mH o>onm HH<

o.om ONe.N :oEEou N
H.om Nwo.m :ossou H
m.eo meo.NH oooHom

ecu meg
HNV

eonHmoHo Hmmv use Nomosou
AHHmsuo< xHHmsuo< kn com:
eHoH» HoHocmso moemoo

 

moHsmom Nesom Hoooo<

monHUHszm mmhzmzou Oh mbqumm rnohm H<DHU< mo zomHm<mzou

n mgm<9 >m<22=w <H<Q wnsbm QHmH> Qz< Hmoo AH“

N >z<mzou

121

.woumHseHm ohoz mowmucoopoa wHoHH HHco ouch wow: mosHm> ommuoom Hmsuum nouns Ho: oc\M

NN.mH we.NH NNN.HN HHS.mH o>oom HH< oHH.mH HHo.NH
HHHm ozHHqu HH<Hm>o
mcoHHeoHaHH oHomo--oHonmoosH :oHomHsaHm oHe.N :oEEou N mmN.H NHN.N
No.9 Hm.HH Noe.m Nwo.m :oEEou H moo.e mmH.e
NH.m Ho.cH eea.mH meo.NH. HooHom+m<m meN.eH moo.He
SOHx oon
HHmS\HHS-HS HHeS\HNS-HS HHS HmmS oeoHo HHS HmmS
mmcH>mm mm mmcH>wm m umou now: :H com: Hmou onHscom
HmHucouom HmHucouom coHHMHseHm \fluonesq owmho ponasq
HSS HHS HeS HmS HNS HHS

muHozu mn<mu Pmoo.%m<mq QmHUHQmmmummhsmzou osz: >m mozH><m H<thmeom

mBHDmmm >Qabm OH monhuHszm awksmzou mo zomHm<mZOu
e mqm<9 >m<zzam <b<m >DDHm QHmH> Qz< Hmou mHm

N >z<mzou

:oEEou N+m<m

cOEEou N+m<u
:oEEou H+m<m
COEEOU N+m<m
HmSoeoHo

onunoum
housano

RIP COST AND YIELD STUDY

122

COMPANY 3

DATA SUMMARY TABLE 1

LUMBER YIELD SUMMARY

 

(1) (2)
Grade Input
(BF)
FAS 378
Select 1247
1 Common 4563
2 Common 3928
Total 10,116
(5) (6)=(3)
Grade Input
(BF)
FAS 280.3
Select 945.2
1 Common 3891.2
2 Common 3455.7
Total 8572.4

(3) (4)
Output Yield (%)
(BF) (3)t(2) x 100
280.3 74.2
945.2 75.8
3891.2 85.3
3.41.512. 88.0
8572.4 84.7
(7) ,(8) (9)
Output Yield (%) Cummulative
(BF) (2)/(6)x100 Yield (%)
(7)/(2)x100
266.4 95.0 70.5
887.4 93.9 71.2
2117.1 54.4 46.4
2153.1 62.3 54.8

5424.0 63.3 53.6

123

---- a.meH n.moH o.eNem mH.Nm m.eH em.m ac.nm HQHOH

 

o.o - m.HmH m.HoH H.mmHN om.mH c.wH eo.m em.oH :oEEou N
5.0m- mHNmH o.emH H.NHHN mN.mH o.eH mN.N mm.mH :oEEou H
N.HH- m.moN H.mHN e.nww mo.e m.e aH.o eN.e NooHom
---- . m.mmN N.mmN e.ooN mo.H H.N wo.c mH.H m<m
owmho HHmSHomS HHmHucouoaS HNSmHmS
omoan: oxoz Hzom\og\mmS Hamm\os\mmS HNS-HNS osHe HHHS HHHS
coozoom 0HNShHoS oomm hHmSmHSS oomm HmmS osHp .ezoa oEHe oeHN
owcmcu m :oHuosuopm :oHuoswon unnuzo poz w czon mmopo owmuo

Hum Hzmuso com: wommmv mzmm mono

---- N.mwm e.oon oHH.oH No.0 H.He CN.e NN.=H Hwooe
N.o - N.oNoH e.HoNH wNam MN.N m.wm Ne.H mo.m :oesou N
H.mm- o.mmoH e.HNoH meme mN.N N.mm we.H HN.e :oeeou H
m.mw+ m.aHoH ”.mmNH NeNH mo.c N.HH mo.o NN.o oooHom
--- N.NmN N.eoa NNm wm.o H.0N HN.H om.H men
. oempo HHmzooaS HHoHocooooS HHHS HNSMHmS
omongz oxoz Hzom\og\mmS Hzmm\oc\mmS HmS-HNS osHH HHHS HHHS
coozoom «HNSIHSS oomm mHmShHSS ooom HmmS oeHe axon oeHH oeHH
omcmgo » :oHuuscoum :oHHostHm usacH Hoz » czom mmOHo owmwu
HNS HNS HNS HOS HmS HeS HNS HNS HHS

Hum psmcH coo: wommmS 3mm mHm mcmo
>m<zzzm mzHe onHuaaomm
N mHm<H >m<223w <h<n >oshm QHmH> oz< Hmoo mHm
m >z<mzoo

124

.mcoHHmHHEHH camcoH :oEEoo N m.Empwoug Hausano owHHHo>o ow wowsHocH w<m NNH >3 coocouscH \M

INmNH N.em ISNSm INm HHNH N.om nemmm coesou N w.em mNmm coeeou N
\H \H \H \H
mNmN H.oe Hmme NS eNNH N.mo omom coEEou H e.oe meme :oEEou H
Nmo o.HN oeNH HSH awe N.ON mNNH oooHom N.HN NeNH oooHom
moH o.oH eoe em HwH N.mN emm we» m.ON wNm we;
HNS HNS HNS . HNS
:Hoooo HNHS o.He< mHH Ham HmmS Hemosou eoonooo HmmS one HomoEou
HHS op one oo eHoHH HHS NooH on use oo Hp eoms NHHoooo< HHHoooo< No eoms
pmou UHoH> quucmao ucmHm Hmou wHoH> szucmso mowmpu vHoH> poucmso mowmuu

 

:oHHmHzeHm xvzum

 

 

mcoHpoHcon Housmsou muHsmom acoum HmSHu<

moneuHDmmm mmeamzoo 0e mhqumm >Qabm H<Dku< mo zomHm<mzoo
m mHm<H >m<zzsm <H<Q >Q=Hm QHmH> Qz< Hwoo mHm

m >z<mzou

125-

.mcoHHmHHEHH camcoH oHan vHon ovHHHo>o 0H wousHucH m<m NNH Ha voocosHmcH xHHmHuuma\M

.woumHaeHm who: mowmucouuon wHoHH HHco ooch com: mosHm> ommuoom Hanuum nouns Ho: oa\w

Nn.wN wm.Hm mHmm ammo o>onm HH< mNom axon m<m
HHHm UZHHHDU HH<mm>o

Ho.oH mN.mN \MNmmH \Mouom :oEEou N NNeH NNmN m<u
mc.em mm.ee mnmN Hmme :oEEou H NNeH HmmN m<u
No.o NH.N ch oeNH HooHom coo GSNH w<m
He.NH NN.o mmH eoe m<m HmH emm m<m

OOHx oOHx
HHmS\HHS-HS HHeS\HNS-HS HHS HmmS ocmHo HHS HmmS HmSoemHo
mwcH>mm mm mmcH>mm a umou wow: :H com: umou onHsvom wouoHuohm
HmHucopom HmHucouod :oHumHssHm .menESH onmuo nonazq pounchu

HOS HmS HeS \ HNS HNS HHS

muHozu mm<mo Hmouohm<mg QmHUHQmmmummHDQEOU osz: >m mozH><m H<thmeom
mHHDmmm renew OH monHuHQmmm zmbamzoo no zomHm<mzou
e mqm<H >m<22=m <H<Q >Qahm QHmH> Qz< Hmoo mHm

m wz<mzou

126
COMPANY 4
RIP COST AND YIELD STUDY DATA SUMMARY TABLE 1

LUMBER YIELD SUMMARY

Cross Cut First Line

Cross Cut and Straight-line Rip Saws

 

Grade INPUt Output Yield (%)
(BF) (BF) (3)5(2) x 100

FAS and Select 6195 4370.8 70.6

1 Common 6000 3636.1 60.6

Total 12,195 8007.4 65.7
(5) (6)=(3) (7) (3)

9
Grade IDPUt Output Yield (%) Cummilgtive
(2)/(6)x100 Yield (%)

(7)/(2)x100

Continuous line-~1nc1uded above

 

 

127

oemuo

HmozmH: axoz
coozuom
owcmnu.»

w.cH-

owmuo

HmogmH: axoz
coozuom
omcmzu »

HmS

N.emH o.mmN e.Ncow mo.mm c.mm
m.NmH N.oeN o.omcm HH.mH o.cm
c.0mH w.mmN m.one em.wH o.mm
HHaoooeS HHaHozooooS HNSnHmS
Hzam\og\mmS Hzam\HH\NHS HmS-HNS osHe
hHNShHoS ooam “HmShHoS ooaz HomS oEHe .ozoo
:oHuosonm :oHHoswoum uzmuso uoz w
Hum psauzo com: wommmS mzmmaHm ocHH-H;mHmHHm
N.mem m.Now mmH.NH CN.mH H.Nm
o.me o.nmn coco No.N N.Nm
H.me o.Hmw mch mN.N m.Hm
HHmsoomS HHaHoeooooS HHHS HNShHmS
Hzmm\og\mmS Hzmm\og\mmS HmS-HNS oeHe
flHNShHSS oomm flHmSmHSS ooom HmmS osHe ozoo
:oHHosvoum :oHHUSUONA usacH uoz N
HNS HNS HSS HmS HeS
Hum HzacH cog: wommmS 3mm usummohu--o:HH HmHHm HDUmmopu

>z<223m mZHb ZOHBUDQozm

N mHm<H >m<22=m <H<Q

 

 

oH.mH mN.Hm Hmuoe

MN.” ew.mN ooseou H

Nm.a Ha.NN oooHom
can m<m

HHHS HHHS

oeHH oeHe

esom mmopu owmgo

MN.H me.NN HmoOH

mN.N HN.HH :oggou H

mm.m co.cH oooHow
ecu mem

HHHS HogS

osHH oeHe

:ZOQ mmopu owmpo

HHS HNS HHS

>QDHm QHmH> Qz< Hmoo mHm

e >z<mzoo

128

eeo.oH H.co meN.HH o>os< HHNm c.Nn

Naoe o.co ”Nam NN Nwoe o.ao
Noam
mmmm o.oN HooH ca NNHH o.mN
HNS HNS HNS
:Hmooo HmmS o.He< NHH Ham
HHS oo use oo eHoH> HHS NOOH om

NHm.oH o>oom HH<
QSNm coeeou H

memm uuoHom
emmH wan m<m

HmmS acmanu
poo ow >3 wow:

Hmoo eHoH> HHHucmso ucmHm “moo wHoH> HHHucmzo mocmpu

 

:oHHmHSEHm vaQm

 

maoHHoHvoum Housmsou

ocHH HmuHm usommouo

monhoHQmmm «mesmzoo Oh mequma >D=Hm.H<DHu< mo zomHm<Azou

m mHm<H >m<223m <H<Q

oHoN
HHS

mmH.NH o>onm HH<

coco

mwNe
NHmH

:oEEou H

HuoHom
flaw m<m

eocHeooo HmmS Hsu Ncoosou
zHHmsuo< zHHmsuo< x3 com:
eHoHH HoHooaoo moeaoo

 

moHsmom Hesom Hmooo<

waahm QHmH> Qz< Hmou mHm

e >z<mzou

.woumHssHm one: mowmucouuon wHoH> >Hco ooch wow: mosHm>

Ne.o Nm.mH CONN NeN.HH o>onm HH<

HHHm OZHHHDU Hq<zm>o

NN.cN mN.Ne meON mwwN :oEEoo <m
o, HN.H o mcnm coma :05500 N
2
l
--- me.H mmmH ommH :OEEou H
HSHSHSH oOHx
HHmS\HHS-HS HHeSNHNS-HS HHS HmmS oomHH
mwcH>mm mm mmcH>mm m umou wow: :H pom:
Hmmucopom HmecoHom :oHumHSEHm \muonESH ovmpo
HcS HHS HHS HNS

ocHH umNHm 3mmmHm mama

owmpoow Hmsuom nouns Ho: oo\m

NNN.cH :oEEou
eamn won :oEEou
eNHH mHHN :oseou
wonm ammo :oEEoo

mom :oEEoo
HHmH HeeH :oEEou
HHS HmmS HmSoemNo
umou vothcom wouunonm

Nonezq Nouznsou
HNS HHS

moHozu mn<zo Hmoo.hm<mq QMHUHcmzm-mmHDQZOU oszD >m mczH><m H<HFzmhom

mbgsmmm >o3Hm OH monHuHQmmm zmhzmzoo no zowH¢<mzou

e mHm<H >m<223m <H<Q WQDHm QHmH> Qz< Hmou mHm

e >z<mzou

HN

130
COMPANY 4
RIP COST AND YIELD STUDY DATA SUMMARY TABLE 1
LUMBER YIELD SUMMARY

 

Gang Rip Saw First Line -- Gang Rip Saw and Cross Cut Saws
(1) (2) (3) (4)
Grade Input Output Yield (%)
(BF) (BF) (3)9(2) x 100
1 Common 1956 1444.4 73.8
2 Common 6906 4684.8 67.8
3A Common 3885 1467.1 50.9
Total 11,747 7596.3 64.7
(5) (6)=(3) (7) (8) (9)
Grade Input Output Yield (%) Cummulative
(2)/(6)x100 Yield (%)
(7)/(2)x100

Continuous line-«included above.

131

N.eN-
N.Nm-
owmuo
omogmHz oxoz
coozuom
omcmzu H

H.oN+
N.oe+

ovmuu

.omosz: oxoz

cooZHom
omcmnu m
HmS

m.NoH

n.on
w.HoH
o.ooH

HHmzpomS
Hzmm\ug\mmS

«HNSIHSS oomm
coHuusvonm

m.owm

H.0wn
w.mNc
c.Nnm

HHmSHUmS
Hzom\og\mmS
.HNSIHSS oomm
:oHHUD©OHm

HmS

m.NmH

m.mm
N.©MH
H.HeN

HHmHucouoaS
H38m\mz\mmS

mHmShHoS oomz

:oHuoswoum

Hum Hampzo cog: wommmS mzmm mono

H.oHOH

c.5aoH
n.eem
e.anH

HHwHucoquS
Hme\»:\mmS
hHmSuHHSS ooom
:oHposonm

HNS

m.camn mo.mm

H.NoeH wc.eH

m.ewoe He.em
e.eeeH mm.m

HHS-HNS

HmmS oeHe
panuso uoz

neu.HH mo.HH

mNNN mo.N
come Hm.N
SmaH mo.H
HHHS

HmS-HNS

HHmS oeHe
usch ooz
HSS HmS

e.mN ON.NH
H.NN oe.v
N.mN am.HH
H.Hm HN.N
HNSIHmS
oeHH HHHS
.fi3om «EMF
w :30Q
8.H4 aN.N
H.NN eo.H
H.Nm ao.m
H.No om.m
HNShHmS
osHH HHHS
£30D OEHF
w c309
HeS HmS

Hum uzacH cog: pommmS 3mmaHm w:mo--o:HH HmHHm 3wmmHm mcmu

>m<223m mZHH ZOHHUDQomm

N mqm<h >m<22=m <H<Q

>QDPm nHmH> Qz< hmoo mHz

e >Z<QZOU

ew.mm HmuoH

eH.mH :oEEoo <m
co.oe :05500 N

on.w flOEEOU H
HES .
osHH
mmouu ovmho
Nm.aH HmHOH

No.m :oEEou em
co.HH :oggou N

mN.m =OEEOU H
HH;S
oeHH

mmOHo owmgu
HNS HHS

 

132

ewHw
mNON
womm

mmmH

HHS

HNS

:Hoooo

0“

ome.HH cNH HHm.oH

omNN
ammo

mmoH

HmmS
you op
Hmou nHoH> szpcmso ucmHm

mmH
oNH

mHH
HNS

o.He<
eHoHH

 

:oHumHseHm xvspm

“NwN
meme

mONH

HHS
Hmoo

m.Hm
m.om
N.mm
e.eo
HNS

wHH Ham
NOOH um
wHoH> Hqucmso mocmpo

moe.eH
Hmmm
ommw

meNN

HmmS
H30 ow

o>on< HH<

:oEEou <m
:05500 N
:oEEou H

>=mgsou
No eoma

 

mcoHuonoNQ Hopsmsou

ocHH HmHHm samem mcmo

m mHm<H >m<22=m <H<Q

e >z<HHZOU

N.eo
a.Om
N.NS
N.mN
HNS

wochuno
AHHmspo< xHHmsHo< kn wow:
uHOH> quucmso mowmho

neN.HH o>on< HH<
mwwN COEEOU <m
ooao COEEOU N

cmmH :oEEou H

HmmS one HoooEou

 

moHsmom Nesow Hmoooe

>Q5Hm QHmH> Qz< hmou mHm

monHuHammm mmHDmZOU OP mhqumm >Qbhm H<Dbu< mo zomHm<mzou

133

.woumHssHm who: mowmucoupom @Hon xHCO ouch vow: mosz> owmuoom Heapom nuume Ho: oa\m

o.N wm.e eeo.oH moH.NH o>onm HH<

o\°

HHHm UZHHHDU HH<mm>o

o o Nmoe coco :oEEou H
mmNe HooHom
Wm.N ww.m mmNm NHmH can m<m
oon oon
HHmSNHHS-H_ HHeS\HNS-HS HHS HmmS ocmHH
mmcH>am mm mwcH>wm m “moo pom: :H com:
HmHHcOHOQ Hmecou0d :oHpmHssHm _\m9onE:H compo
HOS HHS HeS HmS

ocHH umpHm usommONU

MHcm

che

oeme

HHS
umou

HNS

HNN.SH cossou H
NNOH ecu m<m.
mwmm :oEEou H
mmHm :oEEou H
cow. ecu m<m
HmmS HmSoemmo

poNHscom wopunon

honezq h0u3¢Eoo
HHS

moHOIU mm<zo Hm0049m<m4 thuHQmmm-mmHDQZOu qum: >m mozH><m H<Hbzmhom

meqammm waahm OH monHUHmmxm zmhsazoo mo zomHm<mzoo

e mHm<H wa<22=m <H<Q wanem QHmH> Qz< Hmoo mHm

e >z<mzou

134

COMPANY 5

RIP COST AND YIELD STUDY DATA SUMMARY TABLE 1
LUMBER YIELD SUMMARY

Cross Cut Saws

‘(3)

 

(1) (2) (4)
Grade Input Output Yield (%)
(BF) (BF) (3)8(2) x 100

FAS 2163 2357.1 109.0

Select 2880 3094.4 107.4

1 Common 4932 4955.7 100.5

2 Common 3349 3075.6 91.8

3A Common 1898 1453.1 76.6

3B Common 1230 _1§§;1 61-3

Total 16,452 15,689.3 95.4

Rip Saws
(5) (6)=(3) (7) (8) (9)
Grade Input Out ut Yield (%) Cummulative
(BF) (B ) (2)/(6)x100 Yield (8)
(7)/(2)x100

FAS 2357.1 1592.1 67.5 73.6
Select 3094.4 2061.2 66.6 71.6
1 Common 4955.7 2785.7 56.2 56.5
2 Common 3975.6 1611.2 52.4 48.1
3A Common 1453.1 757.4 52.1 39.9
3B Common 753.4 373.4 49.6 30.4
Total 15,689.3 9181.0 58.5 55.8

135

N.mH-
m.m-
w.cN-
m.m-
w.N-

owmhu

Hmosz: axoz
coozpom
omcmsu m

c.c-

e.eN-
w.nm-
e.mN+
N.oH+

. oemoo

HmogmH: onz
coozpom
omcmnu w

HmS

H.NmN

m.mmH
m.wHN
m.mmN
e.NNm
e.mmm
m.mmm

HHwSHomS
Hzam\og\mmS
hHNSIHSS ope:
COHHUD©OHQ

m.mcm

m.mwe
o.on
N.mmo

N.coHH

o.me
N.c@n

HHmspomS
Hamm\mc\mmS
«HNS.HSS oomm
fiOwHUSfiOhm

HNS

N mqm<h >m<22=m <H<Q

m.mmm o.Hme
m.HmH e.mNm
m.mmN e.nmn
m.moN N.HHcH
m.Hom N.mwNN
H.Hmm N.HooN
m.ome H.NmmH
HHmHucouomS
Hzmm\uz\mmS
.HmSIHoS oomm HmmS
:oHHoswopd usapso

o.omm Nme.oH_
N.HNo omNH
N.Nwm mme
H.o.c memm
N.NHNH Nmme
o.mmm ome
m.mom moHN
HHmHucouoaS
szm\gs\mmS
«HmSwHSS ooom HmmS
COHHosooud usncH
HNS HSS

mm.NN

mm.H
mm.N
mm.m
Hn.n
NN.m
on.m

HmS-HNS

oeHH
ooz

Hmm Haauso com: commmS mammaHm

mm.mH

mm.H
mN.m
Hm.e
mo.e
om.N
mm.N

HH:S

HmS-HNS

osHH
uoz

HmS

Hum HSQGH con: pomme mzmm HSUmmoyu

>m<223m mEHb ZOHHUDDomm

m >z<mzou

n.oH

m.o
m.mH
m.NH
N.oH
H.m
e.cH

HNSmHmS

oeHH

«czoa

w

o

c.oH

m.HN
m.HH
H.H
H.m
m.HH
H.NH

HNShHmS

osHH
czoo

H
HeS

>Q3Hm QHmH> Qz< Hmou mHz

mN.

M

No.o
mm.o
ow.o
ww.c
mm.o
me.c

HHHS
oEHe
czoo

mH.N

cm.o
Ne.c
oc.c
He.o
nm.o
mm.o

HoHS
osHH
czon

HHS

em.om

hm.H
ne.m
mw.c
eo.w
om.m
mH.e

HogS
oeHH

mmouo

me.HN

mm.N
mc.m
nw.e
me.e
mN.m
NN.N

HHHS
osHH

mmopo

HNS

Hmooe

:oEEou mm

:OEEou <m

:ossou N

:05500 H
HooHom
m<m

ovmuo

Hmooe

:oEEou mm
:oseou <m
:05500 N
coeeoo H
HuoHom
m<m

compo
HHS

 

136

.mcoHumHHEHH oHan wHoHH ow 03w oHnHmmOQEH :oHHmHDEHm

mmmN e.oo oHee No
NHHN H.NN SHNN SCH
HmmH N.¢N oooN HOH
HNS HNS

:Hmooo HNHS o.He<

HHS oo poo oo eHoH>

umou nHoH> qupcmzo pcmHm

 

:oHumHssHm HwSHm

:oEEou m

:oEEou N
NomN w.mo mNme COEEOU H
emmN N.Hn Heom uuonm
wNoH m.on coHN m<u

HNS
NHH Ham HmmS Hcmoeou
HHS NQOH on 330 oo No eon:
Hmou cHoH> HHHucazo mowmwu

 

mcoHponon hopsmEou

monhuHommm «MHDQZOU OP whqumm >QDHm H<DHU< mo zomHm<mzou
m mqm<e >z<223m <H<Q

m >2<mzou

N.om
HNS

wNHm
memm
Nmoe
cmwN

mcHN

:oaeou m
:oEEou N
:oEEou H

oooHom

m<m

moonoHo HmmS one Homosoo
>HHm5pu< HHstuu< Hp pom:
wHoH> quucmso mowmuu

 

mHHSmom Hanum Hanuo<

>QDHm QHmH> Qz< Hmou mHm

137

.cho :oseoo H ecu .HuoHom .m<m Now wouszou \m
.mcoHHoHHEHH camcoH oHnmu wHoHH ow ozw onHmmoQEH :oHHoHSEHm \M

.coumHseHm opoz mommuaoowon wHoH> HHco ouch com: mosHm> owmpoom Hmzuom nouns Ho: on \M

No.o Nm.w \mmomo Nme.oH o>ooo HH< \mmeNm NHN.eH Hoooe
NH.NH \M \m mNHm ooaeou m ommH omNN :oseoo H
NN.em \m \m memm :oesou N NeNH omHN :oesou H
NN.OH Nc.N mmmN Nmae ooEEou H NomN wNme ooEEou H
o NN.NH NNHN onN HooHom NomH eeHm :oaeou H
o No.H emmH mOHN mem cmeH mmeN coasou H
oon OQHK
HHmSHHHS-HS HHeSNHNS-HS HHS HmmS oooHo HHS HmmS HmSoeoHo
mmcH>om mm mmcH>om a umou pom: :H pom: umoo wohHscom uoHuHcopm
HoHucouom HoHucoHom coHpmHseHm .\muonE:H ocmhu honssq housanu
HoS HmS HeS HmS HNS HHS

moHozu mm<mo Hmou.Hm<MH QmHUHQmmmummeamzou osz: >m mozH><m H<thmhom
msgammm >Qzem OH monhuHQmmm mmhsmzou mo zomHm<mzoo
e mHm<H >m<223m <H<Q >Q=Hm chH> mz< Hmou mHm
m >z<mzou

RIP COST AND YIELD STUDY

138

COMPANY 6

LUMBER YIELD SUMMARY

Cross Cut Saws

(1) (2)
Grade Input
(BF)

FAS and Select 4826

1 Common

2 Common 7223

 

Total

(5) (6)=(3)
Grade Input
(BF)

FAS+Select 4715

1 Common 3763

2 Common

 

Total

 

 

(3) (4)
Output
(BF)
4715 97.7
3763 91.6
6320 87.5
14,798 91.6
Rip Saws
(7) (8)
Output Yield (8)
(BF) (2)/(6)x100
3843 81.5
2959 78.6
4643 73 5
11,445 77.3

DATA SUMMARY TABLE 1

Yield (%)
(3)?(2) x 100

(9)
Cummulative
Yield (8)
(7)/(2)x100
79.6
72.1
64.3

70.8

139

 

in}...

.mHmoa e\e OH New vuozczow ooum3np< \M

---- . o.wme e.mmm scam mN.o c.eH oH.H mN.N Hoooo
m.mm- a.mom H.Nmm (mNo eH.H H.HH Hm.o mc.N coeaoo N
\H
a.OH- o.aoe H.Nmm Nee HN.H H.NH NH.o NN.H :ossoo H
---- . a.on a.NHo aNmN om.m o.eH No.c Ne.4 oooHom
wcm m<m
oemoo HHasoooS HHaHocooooS HNSIHmS
HmoHNH: oxoz. HzNMHNHHHHS Hzam\og\mmS HNS-HNS oeHH HNHS HNHS
coozoom hHNShHHSS ope: 0HmSIHoS ooom HamS oeHe :zoa osHH oeHe
awn—MED w COMHUSUOHQ cowuusthm HSQHSO #02 w G309 mmOhU ofimho

Hum usmuso con: wommmv mzomaHz

---- H.84N H.HHoH HHN.eH NN.N H.8H Ne.H mo.N Hmooo
N.am- (N.moo no.HHN NNNm oo.e H.HH NN.c aN.e coesoo N
\H \H
N.eN- o.Naa N.NemH NoHe HN.H H.8N em.o No.N cossou H
---- a.NHmH m.omeH ONNe No.H o.N mH.o NN.H oooHom
eon we;
. oeoou HHooooov HHoHooooooS HNHS HNSIHMS
omoHNHz oxoz Hzam\og\mmS HaemHNHNNHS HNS-HNS oeHN HHHS HNHS
ooozoom ”HNSIHGS oooz .HmSwHoS oomm HmmS osHH axon oeHe oeHH
owcmsu w :oHpuzwoum :oHHUSpoua usacH uoz N .czoo mmOho owmho
HNS HNS HHS HSS HmS HeS HNS HNS HHS

Hum psacH con: wommmS mzom usommouu
>m<223m mth onhuaoozm
N mqm<e >m<223m <h<n wnahm QHmH» Qz< hmoo mHm

o >z<nzou

140

mcoHumuHEHH oHan wHoH> 0H osw oHnHmmoasH :oHHmHDEHm :OEEou N H.mc mNNN :oEEou N

emwN N.NN cooe oOH HmmN m.mc mmNe :oEEou H N.NN NOHe :oEEou H
mNmm e.ow mmHe OHH mNmfi H.mN NONm wooHom e.ow che pooHom
4 can m<m ecu m<m
HNS HNS HNS HNS
:Hoooo HomS o.He< NHH Hoo HmmS Hzoosou eooHoooo HmmS one Hcoosou
HHS op oou oo eHoHH HHS NooH No one oo No eooa HHHosoo< AHHooooe Ho eoo:
Hmoo oHoH> HuHucmso pcmHm Hmou wHoH> HpHucoso mowmpu nHoH> popcozo.moooho

 

:oHHoHSEHm Hospm

 

 

mcoHuunon Nopsmsou muHSmom Heapm Hosuo<

monHUHammm ambamzou OH mhqsmmz reabm H<=Ho< mo zomHm<mzou
m m4m<9 >m<223m <H<a >asem QHmH> Qz< Hmou mHm

o >z<mzou

 

141

.onoHnaHm ogoz mommucoUNoa oHoHH HHco ooch com: mosHm> omouoow Ho3uuo zooms we: on \m

.mcoHuoHHEHH oHsoH--oHnHmmo;EH

oon

HHmS\HHS-HS
mmcH>mm mm

HmHucouom

HSS

muHozo mm<mo hmou.Hm<mH omkoHommm-mmHszou oszD >m mozH><m H<thmhom

o\°

c.m

o\°

o.m

o\°

H.N

OOHx
HHeSNHNS-HS

mmcH>mm w
HmHucou0d

HmS

emmN

mNmm

HHS
oooo
:oHumfiseHm

HeS

:oHumHSEHm

III-D

\H

mNNn
NoHe

ONme

HmmS
eoo:
uonESH

HmS

:oEEou N

:oEEou H

HuoHom+m<m

ocoHo
:H oom:
osohu

Hume

NQNN

mmwm

HHS
omoo

HNS

Ner

ooam

NNNe

HmmS
wohHscom

honszq

HHS

mhqsmmz >Qaem OH monHUHnmmm mmeamzou no zomHm<mZOU

e mHm<H >m<223w <H<Q

wazkm CHmH> nz< Hmoo mHm

c %z<mzou

owopo><
:oEEou N+H

:05500 H
van m<m

coesou N+H
can m<m

HoSoeoou
wouanoum
howsoEou

142

COMPANY 7
RIP COST AND YIELD STUDY DATA SUMMARY TABLE 1
LUMBER YIELD SUMMARY

Cross Cut Saws

(1) (2) (3) (4)
Grade Input Output Yield (%)
(BF) (BF) (3)3(2) x 100
1 Common 685 686.7 100.2
2 Common 3336 3144.6 94.3
Totals 4021 3831.3 95.3
Rip Saws
(5) (6)=(3) (7) (8) (9)
Grade Input Output Yield (%) Cummulative
(BF) (BF) (2)/(6)x100 Yield (8)
(7)/(2)x100
1 Common 686.7 332.9 48.5 48.6
2 Common 3144.6 1421.4 45.2 42.6

Totals 3831.3 1754.3 45.8 43.6

 

143

m.cN+

ovogo

ooogmH: oxoz.

:oozuom
omcmcu N

N.mm+

ocmpo

.umoan: axoz

:oozuom
omcogu N

HmS

 

 

N.NNH N.oeH m.emNH Ne.NH e.N mH.H No.mH Hoooo
H.mmH N.oeH e.HNeH NN.o o.N oN.c Nm.cH :oEEoe
e.NoH H.HNH o.Nmm mN.N m.HH mm.o H.m oosaou
HHosoomS HHoHoaoooaS HNSmHmS
Hzom\o;\mmS Hzom\m;\omS HmS-HNS oer HNHS HNHS
«HNSnHoS oomz “HmSnHoS ooom HamS oeHe .azoa oaHH oeHe
:oHuuswoum :oHuostHQ “amuse uoz N axon mmopo owmgu
Hum unnuso com: oommmS mzmmaHx
e.Nom a.ooo HNoe NN.m N.NH NN.H mH.H Hmooo
N.oao N.mNN ommm mo.e H.oH NN.o mN.e coEEou
m.maN N.Nom mNo NN.H m.mN oo.o NN.N ooeeoo
HHoooooS HHoHooooooS HNHS HNS.HNS
Hzmm\og\mmS Hzom\o;\amS HNS-HNS .osHo HNHS HHSS
flHNSmHoS ooom .HmSNHoS oomm HmmS oeHe axon oeHN oeHo
:oHuuswoum :oHpostHA NSQCH uoz N :309 mmOpu owouu
HNS HNS HoS HmS HeS HNS HNS HHS

N mqm<H >m<zzam <h<n

Hum usacH con: wommmS mzom usommopu

>z<223m mZHH oneobmomm

N >Z<QEOU

>QDHw QHmH> Dz< hmou QHm

 

144

CNNN H.Nm HNmN ma mNom H.No ooNN ooEEoo N

NmHH e.No mHNH ea ONOH N.NN NHoH :oeaou H
HNS HNS HNS

:Hoooo HmmS o.He< . NHH Ham HmmS Nsmosou

HHS oo poo oo eHoH> HHS NOOH No one oo No mom:

pmoo oHoH> HuHucmao pcoHd umou wHoH> HuHucoso moompo

 

:oHumHzeHm stum

 

mnoHuunohm Nousmsoo

H.Nm ommm :oseou N
m.No mmo coEEou H
HNS

eoonooo HomS one Nnoosoo
HHHosuo< >HHmnuo< >9 wow:
uHoH> zuHucwso moompo

 

moHsmom Nmoom Hosooe

monHUHammm mmhamzou OH mHHDmmm >Q=Pm H<=Hu< mo zomHm<AEOU

m mHm<H >x<zzam <H<Q >QDHm QHmH> Qz< Hmoo mHm

N >z<mzou

145

.couoHJEHm opoz mowoucoouog nHoH> >Hco ooch pom: mosHm> omopoow Hospum gowns uo: on \m

COEEOU N+H

Nm.oN Ne.oc wHee HNoe
oon OOHx
HHmS\HHS-HS HHeS\HNS-HS HHS HamS ocmHo
mmcH>mm mm mwcH>mm N umou vow: :H vow:
HmHucopoa HmHucouom :OHumHsaHm \mponesq owoho
HoS HmS HeS HHS

weNH omHm 205500 N
can m<m
HHS HmmS HoSoeooo
umou wopHscom wouonon
Nonesq Nouzaaou

HNS HHS

muHozu mo<mo Hmoo.9m<m4 QmHoHQmmmummhzmzou osz: >m mozH><m H<thmkom

mhqnmmm waaHm OH monhuHQmmm dubsmzoo mo zomHz<mzou

e mHm<H >m<223m <H<Q
N wz<mzou

>Qzem :HmH> Qz< Hmou mHm

146
COMPANY 8
RIP COST AND YIELD STUDY DATA SUMMARY TABLE 1
LUMBER YIELD SUMMARY

Cross Cut Saws

(1) (2) (3) (4)
Grade Input Output Yield (8)
(BF) (BF) (3):(2) x 100
Select 1700 1667.2 98.1
1 Common 6374 6278.3 98.5
2 Common 931 868.8 93.3
Total 9005 8814.3 97.9
Rip Saws
(5) (6)=(3) (7) (8) (9) '
Grade Input Output Yield (8) Cummulative
(BF) (BF) (2)/(6)x100 Yield (8)
(7)/(2)X100
Select 1139.1 597.7 52.5 54.4
1 Common 4743.6 3292.9 67.5 69.0
2 Common 887.4 319.5 36.0 42.7

Total 6770.1 4120.1 - 60.9 ' 63.0

147

N.HH+

c.mN-

omega

amoan: axoz
:oozoom
oucmgu N

m.mm-
e.N+

owmuo

.omoHNH: oxoz

:ooZuom,
omcmgu N

HmS

H.ew

N.mmH
w.mNH
w.mNN

HHmzpooS
Hzom\pz\mmv
«HNSmHoS ooem
:oHuusuoum

m.ome
m.wmm
m.Nom
o.mme
HHaspooS
HaemHNeNmmS.

hHNSmHoS ooma
:oHuosvoum

HNS

N mqm<B >m<223m <H<Q

e.HHH H.oNHe oo.nm m.eN Ho.NH No.oe Hmuoe
m.NeN m.mHm ew.m m.oN om.o Nw.e 205500 N
H.mNN m.Non. um.wN H.NN me.m No.0m :oEsoo H
e.omm 5.5mm mm.e o.mm eo.N mm.N pooHom
HHeHoeooooS HNSCHNS
HaemNNHNLNS HmS-HNS oeHe HeeS HoeS
mHmSmHeS ooem HmmS oeHe .ezoa oeHe oeHe
:oHuosoopm undone uoz N czom mmopo oompu
Ham usmuso :oa: nommmS mzmmaHm
m.aHN moom Nm.NH N.mm NN.o eN.mH Hmuoe
H.Nec Hmm me.H m.Ne om.H mN.N :oEEou N
N.NON enmc oo.m m.wN om.m om.NH coesoo H
m.HNm ooNH no.N N.mm ow.H me.m pooHom
HHoHoeooooS HoeS HNSmHmS
Hzom\oe\mmS HNS-HNS oeHe HneS HNHS
IHmSIHoS oomm HmmS oeHe axon oeHe oeHe
:oHposzHA usncH uoz N. czom mmopo owmho
HNS HoS HmS HeS HNS HNS . HHS

Hum usch :oa: womomS mzmm NaommOHU

>m<223m mZHH ZOHHUDQoma
>QDhm nHmH> nz< hmou mHm
w >z<mzoo

148

.mcoHumpHEHH oHnmu wHon on osv

emoN o.No oNoe No
eNm N.Nm NNHH NN
HNS HNS

eHeoeo HmHS o_He<

HHS oo poo oo eHoH>

omoo eHoH> HoHoeeoo oeeHa

 

eoHoeHseHm Nezom

oHnHmmanH :oHHmHBEHm :05500 N

HooN o.mo Name :05500 H

oHe m.HN cow oooHom
HNS

mHH Ham HmmS xcmasou

HHS NSOH om you op Np eom:

umoo oHoH> quucoso mowmho

 

mcoHponopm Nopsmsou

em HNo coseoo N
m.Ne eNmo :oeaoo H
m.Nm OONH oooHom
HNS

eochono HmmS “so Nemoeou
xHHospu< xHHmsuo< Hp cow:
eHoH> NoHoeeso moeomu

 

moHsmom Nooom Hmzooe

wonhuHDmmm awhamzoo OH mHHDmmm rashm H<DHU< mo zomHm<mEOU

m mqm<b >m<zzzm <H<Q

w >z<mzou

>QDHm QHmH> Qz< Hmou mHm

149

 

.cho :oEEoo H wco uooHom pom oopsaeou \m
.mcoHumuHEHH oHnmp oHoHH ow osc oHnHmmonEH oHnHmmodEH :oHumHseHm \M

.eonHSEHm ohoz momoucouuon wHon HHco ooch wow: mosHo> owmuoom Hosuuo nouns No: on \M

:oEEou H
NN.mH Nm.Nm NNNm eNow + oooHom NmNN mNee \mHoooe
Nm.mm \m \m HNm eoeeou N moN mom wee
NN.em NN.HH emoN eNmo eossou H NmmN eoHe mem
NN.Hm No.mm eNm OONH oooHom HNm oHN m<m
OOHx oon
HHmSNHHS-HS HHeS\HNS-HS HNS HaaS oeoHa HHS HmmS HmSoeooo
mmcH>om mm mwcH>om N umou vow: :H com: umou wouHscom voHoHvopa
HmHucooom HmHucouom :oHuoHaaHm \maonssq oompo honszq houzmeou
HoS HmS HeS HNS HNS HHS

muHOIU ma<mo Hmouokm<m4 awhuHammmuamHDQZOU oszD >m mozH><m H<Hbzmkom
mHHDmmm >QDPm 0% monPoHszm zmkamzou mo zomHz<mzou
e mqm<k >m<22=m <H<n ¥Qakm QHmH> QZ< hmou mHa

m >z<mzou

150

COMPANY 9
RIP COST AND YIELD STUDY DATA SUMMARY TABLE 1
LUMBER YIELD SUMMARY

Gang Rip Saw First Line -- Gang Rip Saw and Cross Cut Saw
(1) (2) (3) (4)
Grade Input Output Yield (8)
(BF) (BF) (3)3(2) x 100
FAS 1807 1056.4 58.5
Select 2636 1185.3 45.0
1 Common 3196 1488.6 46.6
2 Common 1194 497.7 40.1
Total 8833 4228.0 47.9
(5) (6)=(3) (7) (8) (9)
Grade Input Output Yield (8) Cummulative
(2)/(6)x100 Yield (8)
(7)/(2)x100

Continuous line--inc1uded above.

 

 

151

---- e.NOH H.NNH o.NNNe me.mN H.Ne NN.NH NN.oe Heooe
e.oN- o.HN o.HmH N.Noe ON.N m.me mH.m mo.o eoeeou N
N.om- N.No H.NoH o.NNeH Ha.N N.He am.e ON.mH eoeeou H
m.om+ e.NmH a.NmH N.mNHH me.m e.mN NN.H mN.N oooHom
---- e.No N.HHN e.omOH ee.e o.em oN.m mN.OH mea
oeeoo HHesooeS HHeHoeooooS HNSmHmS

oooeNHz oxoz. Hzem\oe\mmS Hzem\o;\amS HNS-HNS oeHe HomS HoeS
:oozoom «HNS.HoS ooma “HmSmHoS oped HmHS oeHe axon oeHH oeHe
owcmnu N :oHuusvopm :oHpoaeoum psnpso uoz N czoa mmopo owmuo
Hum psauso con: wommmS mzmm mono
---- e.oem N.Nem mmNN NH.oH N.om oH.o mN.NN Hoooe
m.NN- N.on m.oNe emHH me.N H.HN No.o NH.N eoEEou N
o.Nm- e.moN e.eme oeHm eo.N o.em mN.N NN.¢H eoeeou H
o.Nm+ N.oNe H.Neo omoN oo.e o.oN Ne.H mm.m oooHom
---- N.mHm H.HHN NONH Nm.N m.om eN.m oN.m mea
. oeemu HHooooeS HHeHoeooooS HmeS HNSnHmS
omoeNH: oxoz Hzem\oe\mmS HzomNNHNmmS HNS-HNS oeHe HNHS HNHS
eoozoom mHNSoHoS ooeg *HmSmHoS oomm HomS oeHe axon oeHe oeHe
omcmzu N :oHuuzwopm :oHuoseOHQ psacH uoz N czoo mmoao owmho
HNS HNS HNS HoS HmS HeS HNS HNS HHS

Hum usacH con: wommmS mzom usommONU
>m<zzam mzHe onhuanomm
N mHm<P >z<zzam <H<a >kam QHmH> Qz< Hmou mHm

m >z<mzou

152

Hmm
mNOH
mowH

NemH

HNS

N.wm

HNS

:Hmooo

OH

mNNH
NHcH
HmoN

mme

HmmS
one op
umou oHoH> zuHucozo ucon

Nu
on
mm

mm

HNS
o.He<
eHoH>

 

:oHumHssHm Modem

oem
an
eomH

HwHH

HNS

e.om
o.oo
o.Hc
N.oo

HNS

wHH 4mm
NOOH um

Nwm
oNHH
ONmH

mHoH

HmmS
one op

:oEEou N
:06500 H
uuoHom

m<m

zcomsou
Ne eoms

umou wHoH> qupcoso mowopo

 

mcoHuonoum popsmEoo

m mqm<b >m<223m <H<Q

m >z<mzou

H.ce
c.oe
o.me
m.wm

HNS

emHH
omHm
omON

nowH

:OEEou N
coEEou H
pooHom

m<m

eoeHmono HmmS poo Nemosou
zHHoauo< zHHonpu< >2 com:
wHoH> quucmzo mooouo

 

moHsmom Nesom Hozoo<

monHUHnmmm mmHDQEOU OP mHHDmmm >QDFm H<DHU< mo zomHm<mzou
wnbhm QHmH> nz< hmou me

153

.zpwcoH How mcoHHmuHEHH oHnou wHon m.pou:nEoo ooHpHo>o 0H :oEEoo H Nm moosHocH \M

.woumHseHm opoz mommpcoUNoa oHoHH cho ooch wow: mosHm> ommuoom Hmsuum :uuoe no: on \M

meme New eoeeou H
No.mN No.NN Neoe come o>oeo HH< momm cow NoH we;
HHHH ozHeeao HH<Nm>o
Ne.mm No.6m HmH NNNH NNeoEEou N oNe omN eoeeou H.
No.cm No.om mNoH NHoH seasou H NHN aNHH eoeeou H
NNe eoeeou H
NN.HN NH.em meNH HmoN oooHom oomH HcaH Nmm. we;
No.HH No.NH NemH mmNH mem HNHH HmeH wee
OOHx oon
HHmS\HHS-HH HHeS\HNS-HS HHS HmmS oeeHa HHS HmmS HmSoeeHo
mwcH>mm mm mwcH>om N umou mom: :H com: umoo vohHscom wouoHuoum
HmHucoHoa HoHHCoHom :oHucHseHm \muonesq oompo Nonasq Nouzasou
HoS HmS HeS HNS HNS HHS

muHozu mm<zu Hmoumkm<mq omHUHszmummHDQZOU osz: >m mo2H><m H<HFszom
mhqumm reabm OH monHUHammm mmhzmzou mo zomHm<mzou
e mHm<H >m<22=m <H<Q wcafim QHmH> Qz< Hmou mHm

m >Z<LZOU

154

COMPANY 10
RIP COST AND YIELD STUDY DATA SUMMARY TABLE 1
LUMBER YIELD SUMMARY

Cross Cut First Plant -- Cross Cut Saws
(1) (2) (3) (4)
Grade Input Output Yield (8)
(BF) (BF) (3)9(2) x 100
2 Common 3912 2786.8 71.2
3 Common 3991 1971.3 49.4
Total 7903 4758.1 60.2
Rip Saws
(5) (6)=(3) (7) (8) (9)
Grade Input Out ut Yield (8) Cummulative
(BF) (B ) (2)/(6)x100 Yield (8)
(7)/(2)x100
2 Common 2786.8 1893.7 68.0 48.4
3 Common 1971.3 1380.7 70.0 34.6

Total 4758.1 3274.4 68.8 41.4

155

H.0H-

ocoho

NmocwH: onz.

:oozpom
omconu N

e.wN-

ooopm

.umost: uxoz

coozuom
owco:o N

HmS

me.Hm
mN.aN
Hm.wm
HHozuomS
HeemHNHNNHS

hHNSiHoS ooem
:oHuoswopm

No.00e
mH.moe

mN.Ncm

HHmsuomS
HBmm\p:\mmS
mHNSmHoS ooea
:oHuusoopm

HNS

Hum usacH con: womomS mzmm «summopu --

N mqm<9 >m<zzam <H<Q

HH.mHH m.NNHN mN.NH
No.NoH o.NHN mo.N
mN.NNH m.eoeH No.HH
HHmHucouogS
Hzem\m;\amS HNS-HNS
.HmS.HeS oooe HmHS oeHe
poopso uoz

:oHpuswopm

Hum undone com: wommmS mzmmaHm

Ho.Noo mooN NN.HH
Hm.moo Hmmm mm.o
NN.NeN NHNN NN.m
HHoHoeooooS HNHS

Hzom\ne\amS HmS-HNS
.HmSoHoS ooox HmmS oeHe
:oHuosoONm unacH uoz
HNS HoS HmS

>m<EZDm mZHH ZOHHUDQomm

OH >z<mzou

o.ON
a.HN
w.mH

HNSNHNS

oeHe
.czoa

m.

N.eN

HNSNHNS

oeHe
czon

N
HeS

>QDHw aHmH> Qz< Hmou mHm

HmcS
oeHe
czoo

HNHS
oEHH
c309

HmS

ocHH pmpHm uzommopu

um.mN Hmuoe

o:.m coeeou

mm . HNH COEEOU
HNHS
oeHN

mmohu ocopu
mw.oH Hmpoe
om . m COEEOU
mm.c COEEou
HNHS
oeHN

mmONo ovopo
HNS HHS

156

 

eNme e.He omNH o>oo< eeNm m.em NNom o>oo<

NemN m.em aaam ea HNNH N.oe omoN eoeeou m

ameN N.Ne mawm NN NHNH N.Ho Nmom seasou N
HNS HNS HNS

:Hmooo HmmS H.He< NHH Ham HamS N:oosoo

HHS oH one oo eHoH» HHS NooH oe one oo No eoma

omou eHoHN NoHoeeoo oeoHa

 

:oHumstHm Nonpm

umou wHoH> szpcoso mopmho

 

mcoHuonopm housgaoo

ocHH umHHm usommopu

e.He moaN o>oo<
o.em Hmmm coEEoo m
e.Ne NHmm :oseoo N
HNS

oocHouno HmmS you mamasou
AHHmzpu< NHHospo< Np mom:
nHoH> NuHucoso mouohu

 

moHsmom Nmoom Heoooe

monhoHQmmm mmfiamzou Oh mHHDmmm >bew H<DHU< mo zomHm<mzou

m mum<b >m<22=m <B<Q

OH >z<mzou

>bem QHmH> az< Hmoo mHm

157

.ooumHseHm ouoz mommpcouyom wHon NHco ouch wow: mosHm> ommuoow Hosuum :ouoe No: on \M

NH.ON Ne.Ne emoe moon :OEEOU m+N mNmm . emme COEEOU fl

HHHm UZHHHDU HH<mm>O

No.HH Ne.ee NemN Hmam eoeeou m eHeH oeNN eoeeou N
NN.Nm NN.o omeN NHam eoEEou N HaNN NNeN eoeeou H
OOHx oon

HHNSNHHS-H. HHeSNHNS-HS HHS HmmS oeoHa HHS HmmS HmSoeemo

mmcH>mm mm mmcH>mm N umoo vow: :H mom: umoo ooNstom wouoHvopm

HoHucouom Hoszouom :oHumHoEHm \Muonssq oumho honESH housanu
HoS HNS HeS HmS HNS HHS

ocHH HmHHm usummONu

mUHOIu mm<zo Hmoumhm<mH othHommm-mthQZOu Usz: >m mozH><m H<szmeom
mbqumm >Qaem OH monhonmmm mmHDQEOU mo zomHm<mZOU
e m4m<b >m<22=m <H<Q >thm :HmH> mz< Hmou mHm

oH >z<mzou

158

COMPANY 10
RIP COST AND YIELD STUDY DATA SUMMARY TABLE 1
LUMBER YIELD SUMMARY

Rip First Plant

(1) (2) (3) (4)
Grade Input Output Yield (8)
(BF) (BF) (3)5(2) x 100

Gang Rip Saw
1 Common 5101 4362 85.5

Straight-line Rip Saw

1 Common 4900 4394 89.7
Total 10,001 8756 87.6
Chop Saws
(5) (6)=(3) (7) (8) (9)
Grade In ut Out ut Yield (8) Cummulative
(B ) (B ) (2)/(6)x100 Yield (8)
1/ (7)/(2)x100
1 Common 6843— 4963 72.5 63.5 2/

1/ 1913 Board feet of 4 inch or random width was not chopped.

2/ Adjusted for rippings not chopped.

159

oompo

amoan: axoz_
:oozuom
oncogu N

. oomuu
Hmoan: uxoz
sooZNom
omcmzu N
HmS

 

:m.NmH oe.nmN meme om.oN N.mm Hc.cH
HHeeooeS HHeHoeoooeS HNS.HNS
HeemHNeNmmS HeemNNHNmNS HHS-HNS oeHe HoeS
«HNSIHoS ooem «HmSmHoS ooem HomS oeHe .ezom oeHe
:oHuuswoum :oHuozoONm “amuse uoz N czom
Hum unquso com: pomomS mzmm mono
e.ooo o.mmm some on.m e.mN Nm.N
Hum usacH con: wommmS 3mmgHm ocHH-p:mHmpum
N.mCN N.meHH Hon oe.e m.wm mN.N
Hum usmcH com: womomS 3mmnHm mcou
HHesooeS HHeHoeooooS HNHS HNSmHmS
Hzom\ee\mmS Hzom\me\mmS HmS-HNS oeHe HmeS
0HNSmHoS oomm .HmSmHoS ooea HmmS oeHe ezoa oeHe
:oHpu:©0Hm :oHuostHm psacH uoz N czoa
HNS HHS HoS HmS HeS HNS

oeHH ommHm on

>m<zzaw mEHH ZOHHUDQomm

N mHm<H >m<zzsm <H<Q
oH >z<m200

>QDHw QHmH> Qz< Fmou mHm

Hm.Hm

HNHS
oeHe
mmohu

me.m

mN.N

ngS
oeHe
mmONo

HNS

:oEEou H

onouo

:05500 H

:oEEou H

oompo

HHS

O
/O
11
Hoem N.mo eHNo Na mmem N.mo NHmo :oeeoo H
HNS HNS HNS
:Hmooo HmmS o.He< NHH Ham HmmS Neeoeou
HHS oo one op eHoHN HHS NOOH om one oH Ne woo:

umou oHoH> poucoso ucon pmou wHoH> quucoso mocmuo

 

:oHuoHseHm Nozpm

mcoHHoHooum Nousmsou

oeHH omnHm on

m.mo memo COEEOU H

HNS
wocHouno HmmS you xcmosou
zHHm3uo< zHHmsuo< kn wow:
eHoH> NoHocmso moeemo

 

moHsmom Neoom Hmsooe

monHuHommm mmksmzou Oh mHHDmmm >QDHm H<DHU< mo zomHm<mZoo

m mHm<H wd<223m <H<Q

oH >z<mzou

>QDHm QHmH> Qz< Hmoo mHm

161

.wopmHsaHm ohoz mowoucounoa wHoH> cho ooch com: mosHm> ommpoom Hanuuo gonna go: on \m

:oEEoo N
mam

--- NN.NH Hoom meme eoEEou H eNNm mHmN eoEEou H

one mem

OOHx OOHx
HHmSHHHS-HS HHeS\HNS-HS HNS HmmS oeeHo HHS HamS HmSoeoNo
meHHHmm mm mwcw>mm a HmOU 60m: Hz... fimmD HmOU fimszcmm Umuuflwmpm
Hmfiufiouom Hawpfiouom COHumHDEMw \Im. Hons—SA owmho hon—EDA .HGHZQEOU
HoS HmS HeS HNS HNS HHS

oeHH ommHm on
muHozo mnemo emoo.em<mH ameoHamao-mme=ozou quma >m mozH><m H<Hezmeoo
meHammm Noaem ON monNanmma zmezazoo mo zomeeazou
e mHm<e Nmezzam eeea >asem chH» mze emou on

oH >z<mzou

LIST OF REFERENCES

 

LIST OF REFERENCES

Ross, Vincent R., Unpublished presentation at RIP (Rough-
mill Improvement Program) Training Session, Orlando,
Florida, 1978.

Vaughn, C.L., A.C. Wollin, K.A. McDonald, and E.H.
Hardwood Longrades for Standard Lumber. USDA Forest
Service Research Paper, FPL 63, Madison, Wisconsin
June, 1966.

Englerth, George H., Charts for Calculating Dimension Yields
from.Hard Maple Lumber. USDA Forest Service Re-
search Paper FPL 118, 1969.

Moser, H.C., "Potentials for Increased Profits Through Analysis
of Yields and Operational Procedures." Rough Mill
Yield and Operations Seminar Proceedings, North Caro-
lina State University, Raleigh, N.C., December, 1966,

7 pages.

Ross, Vincent R., "Management's Roles in a Program for Yield
Optimization," Rough Mill Yield and Operations
Seminar Proceedings, North Carolina State University
Raleigh, N.C., December 1966, 29-33.

Thomas, R.J., The Yield of Dimension Stock, Volumes 1, 2 and
3. Number of Maximum Cuttings per 1000 Board Feet.
North Carolina State University Technical Bulletin,
No. 24 A, Raleigh, N.C., 1956.

French, Don., "Here' 3 How Our Rough Mill Yield Program Pays
Dividends," Wood and Wood Products, 70 (May 1965
27-29, 105.

Keppler, W.E. and R.J. Thomas., "New Two-way System Predicts
Lumber Yield." Wood and Wood Products, 70. (October

Keppler, W. E. and R. J. Thomas. "How to Predict Costs by

Using New Yield Data." Wood and Wood Products, 70
(November, 1965), 25, 69- ~70.

162

 

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

163

,Englerth, George H., and Daniel E. Dunmire, "Program-

ming for Lumber Yield." Forest Products Journal, 16
(October, 1966).

Wodzinski, Claudia and Eldona Hahm. A Computer Program
to Determine Yields of Lumber. Forest Products Lab-
oratory Unnumbered Report, March 1966.

Dunmire, Daniel E. and George H. Englerth. Development
of a Computer Method for Predicting Lumber Cutting
Yields. USDA Forest Service Research Paper NC-15,

1967.

Schumann, David R. and George H. Englerth. Yields of
Random-Width Dimension from 4/4 Hard Maple Lumber.
USDA Forest Service Research Paper FPL 81, September

1967.

Schumann, David R. and George H. Englerth. Dimension
Stock — Yields of Specific Width Cuttings from 414
Hard Maple Lumber. USDA Forest Service Research
Paper FPL 85, December 1967.

 

Schumann, David R., Dimension Yields from Black Walnut
Lumber. USDA Forest Service Research Paper FPL 162,
1971.

Schumann, David R., Dimension Yields from Alder Lumber.
USDA Forest Service Research Paper FPL 170, 1972.

White, C.H., "Walnut Yields." American Walnut Manu-
facturers Association, Chicago, Illinois, 1950.

Creighton, J.W., W.G. Stump, and W.F. Hutchins, "Cor-
relation of Walnut Furniture Cutting Requirements
with Grade Yield" Mighigan State University,
Qparterly Bulletin, 357(November, 1952), 230-247.

Anonymous. "New Yield Data for Cutting Hardwood Dimen-
sion Stock." Wood and Wood Products, 73, (March,
1968). 31-32, 66.

Schumann, David R. and Henry A. Huber. "Comparison of
Predicted and Actual Yields of Dimension in the
Grand Rapids, Michigan Area."

.Manufiaaiuning, 41 (1969), 54- 62.

Huber, Henry A. and George Vasiliou. "Cost Analysis
in Wood Products Manufacturing." COOperative Ex-
tension Service, Michigan State University, East
Lansing, Michigan, 1968.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

164

Huber, Henry A., "In the Rough Mill Should You Rip or
Crosscut First?" Woodworkingiand Furniture Digest,
26 (June, 1974), 48-50.

Huber, Henry A. and Steven B. Harsh, "Rough-Mill Im-
provement Program." Woodworking and Furniture Digest,

19 (February, 1977, 4 pages).

Anonymous. "Computer Program Yields Lumber Saving."
Furniture Production Magazine, 4 (October, 1977),
25.

Anonymous. "Program Takes Guesswork out of Hardwood
Lumber Cutting." Southern Lumberman, (February,

1978). 17.

Huber, Henry A., Stephen B. Harsh, and Ed Pepke. "Im-
proving Lumber Yields in the Rough Mill." Wood and
Wood Products, 83, (April, 1978), 37-38.

Rules for the Measurement and Inspection of Hardwood
and Cypress Lumber. National Hardwood Lumber As-
sociation, Chicago, Illinois, 1974.

Bullard, Gordon. Telephone conversation December 19,
1979 with author.

Rules for the Measurement and Inspection of Hardwood
and Cypress Lumber. National Hardwood Lumber As-
sociation, Chicago, Illinois, 1974, 113.

Rules for the Measurement and Inspection of Hardwood
and Cypress Lumber. National Hardwood Lumber As-
sociation, Chicago, Illinois, 1974, 107.

 

Harsh, Stephen, Henry Huber, Ed Pepke and Paula Johnson.
Optimum Furniture Cuttinngrogram,iA Teleplan Program,
User's Manual. Department of Agricultural Economics,
Michigan State University, East Lansing, Mi, 1977.

Pepke, Ed and Michael J. Kroon. Rough Mill Operators
Guide to Better Cutting Practices. USDA Forest Ser-
vice, Northeastern Area State and Private Forestry,
St. Paul, Minnesota, 1980.

Mundel, Marvin E. Motion and Time Study; Improving
Productivity, Fifth Edition, New Jersey, PrentiCe-
Hall, Inc. 1978.

 

Meynard, Harold Bright. Industrial Engineering Hand-
book. Third Edition, New York, McGraw-Hill Co., 1971.

165

35. Carroll, Phil Jr. Timestudyfifor Cost Control, Second
Edition, New York, McGraw-Hill Co., 1943.

36. Wood Handbook: Wood as an Engineering_Materia1. USDA
Agriculture Handbook No. 72, USDA Forest Service,
1974.

uniu(All)(W)

1))

W

IIIIIHWI

1293 03142 8

3