1 -21-
{1.12“}:fll‘fi
.

J a. .'~.
'~ 'f;,n'«' «wit:
:fy"1"r. .1“ “.- 1‘ti" ’-.- " . 7.}. V . A "N
‘ ' - .“r f . 5 4.. <

“ V1.13;
" 3:3?“

:s
.135... -'.J. . .. .
up...“ 1.317;.—
u:

:5...

‘
“mglizt’ ‘1
‘35}0‘ u

V;§kfl‘f\ 22%;};

“17¢“wa

Eta-.53

".- -.u-t§ .1.

Q“
a
, "1‘“’J’E‘L‘E‘3

$1
'4

‘E"
LI
&'

I

m.-.....\- 75:...

‘ . pfi'c-RK ‘W- 1‘.

-~‘0 $3.53.

- Squ’ui 1"“:

'5“: .37..-. m .
”“51 u .«v Z.” ‘Y"'

‘3‘ ”NIH

:

"274.4%... .31...-
~ -. ..
:5“. "'5...

‘ ., 111,.
"1‘ ’ 1 A

y
if

a ~‘n1-
~11: - w m '
Q -.. ‘ .J. 73-; 1i",
Qua-4:“ . ~ «24': . " r 1 4 v ‘ - I
“N: fwd-luvs «. .1 " ' ' 1" 1' 'M I iv; 1 m :3‘ "“ 1” “h
‘ . ' “if?“ III?! akmmuvm
.1" ".41 31 . m o ( «'51:?

2-1.;

a
if.
1

I.

i;-’4'-‘u 3"
h! 1. “iguana
mu-

1
. .4
1-5:... ‘1'

.1“ 3,131.3
..

L“n““
N.

' .‘a’x

‘-

,
.V I:
‘n‘l,

' “ " 1:3:

«31111

. . .301.
vu’lk'. 11.11.11

4-1 :-
..-
‘ . . "."’.‘.TI.‘-'.-1‘."""‘:."
".'i.""-‘»'4’.‘. " " ‘-" ' 71-"

..,
,... r
I}. ”fail

.m

 

 

 

Tklé 31‘ D

[V RSITY UBRARIES

.6... llllllll ll

 

 

 

 

 

 

 

 

 

 

 

 

l l l llil'll l l

"'1
M293 00900 8859 .—.

Illll

This is to certify that the
thesis entitled
MANAGERIAL METHODS OF CONTROLLING
PRODUCT QUALITY IN DISTRIBUTION:
DAMAGE MEASUREMENT, ANALYSIS AND CORRECTION

presented by

CARIS JEAN PALMER

has been accepted towards fulfillment
of the requirements for

MS degree in PACKAGING

 

 

JM

Major professor

Date April 30, 1991

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

-—_—___ v

__ __\,
L£BRARY
Michigan State

1 University

LL A

fl

 

 

PLACE IN RETURN BOX to remove We checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunlty Institution
ammo-m

MANAGERIAL METHODS OF CONTROLLING
PRODUCT QUALITY IN DISTRIBUTION:
DAMAGE MEASUREMENT, ANALYSIS, AND CORRECTION

BY

Caris Jean Palmer

A THESIS

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

MASTER OF SCIENCE

School of Packaging

1991

&\

@57—3

ABSTRACT
MANAGERIAL METHODS OF CONTROLLING

PRODUCT QUALITY IN DISTRIBUTION:
DAMAGE MEASUREMENT, ANALYSIS, AND CORRECTION

BY

Caris Jean Palmer

Although firms may estimate how much and what kind of damage
their products experience after they leave the factory, most
do not have a formalized system of managing their distribution
damage. This thesis examines existing management.methods for
maintaining the quality of products in distribution, based on
the analysis of five companies in the office furniture
industry. It recommends developing a systematic approach to
controlling quality, by setting performance standards,
tracking and analyzing conformance, and then upgrading
performance levels and reevaluating standards. It also
suggests potential benefits of managing damage information,
including a reduction in damage, decreases in replacement and

overpackaging costs, and improved customer service.

Dedicated to
my best friend Brian.

ii

ACKNOWLEDGEMENTS

I would like to express my gratitude to the members of my
committee: Chairperson Dr. Diana Twede and Dr. Gary Burgess,
of the School of Packaging, and Dr. David Closs and Dr. Lloyd
Rinehart of the School of Business, Department of Marketing
and Transportation Administration. I would also like to thank
the participating firms, and the representatives who gave

their time and input.
Most importantly, I would like to thank the people who helped

me .complete my degree(s) by offering both financial and

emotional support -- my family and friends.

iii

TABLE OF CONTENTS

CHAPTER I - Introduction .
PURPOSE . . . . . . .
Packaging . . .
CLARIFICATIONS . . .
Transit Claims

Types of Damage

Loss and Damage

CHAPTER II - Literature Review
TOTAL QUALITY CONTROL . .

THE MODEL . . . . . . . .
PRODUCTION QUALITY . . . .
Setting Standards . .
Monitoring the Process
Analyzing Conformance

Corrective Action and Planned

QUALITY MODEL APPLICATIONS

QUALITY IN DISTRIBUTION .
Setting the Standards
Monitoring the Process

Management of Information

Analyzing Conformance

Corrective Action and Planned

CHAPTER III - Research Method and

THE CASE RESEARCH METHOD .
THE CASE SAMPLES . . . . .
THE METHOD OF ANALYSIS . .

CHAPTER IV - Results . . . . .
SETTING STANDARDS . . . .
MONITORING THE PROCESS . .
ANALYZING CONFORMANCE . .

DISCUSSION . . . . . . . .

QUALITY MODEL APPLICATIONS

CORRECTIVE ACTION AND PLANNED IMPR

CHAPTER V - Problems and Recommendations

RECOMMENDATIONS . . . .
BENEFITS . . . . . . . .
PACKAGE IMPLICATIONS . .
IN CONCLUSION . . . . .

iv

Firms

2....
E

o o o 0 Ho 0 o o o o o

t
t

Improvemen s

Introduced

mprovemen s

H
HkaO \lxlONONOKUlH

H
w

REFERENCES .
APPENDICES .
APPENDIX
APPENDIX
APPENDIX
APPENDIX
APPENDIX

MUOCDIP“ °

75
78
78
82
84
87
90

CHAPTER I

INTRODUCTION: THE DAMAGE PROBLEM

Many U.S. firms adopted a new type of "total quality control"
system in the 1980's. The trend toward assuring total
organizational quality was in response to consumer demands for
more reliable products and better service. The success of
firms with effective quality strategies is well documented.
"The return-on-investment from strong and effective quality
programs is providing excellent profitability results . .
.demonstrated.by substantial increases in market penetration,
by major improvements in total productivity, by much lower
costs of quality, and by stronger competitive leadership"

(Feigenbaum 1988, xxi).

Unfortunately, most companies do not continue to apply these
same quality philosophies to their products once they are in
the distribution channels. The "control" seems to end when
the products leave the factory, and it is typically not known
if quality is maintained unless customer complaints are

received.

2
When discussing the many leading companies that have
implemented quality processes in production of goods and
services, Bingham.states that "the<challenge of the '90'8'Will
be to extend this process into the logistics and distribution
functions" (1989) . Indeed, the theme for the 1990 Annual
Conference of the Council of Logistics Management was
"Logistics Quality." The topics ranged from logistics costs
to improving services, but no presentations were offered about
damage or control of product quality in distribution. The
only sessions discussing defects referred to incoming
materials or goods in production, not to quality of products

after leaving the manufacturer.

A manufacturing firm's distribution channels are made up of
wholesalers and retailers who each take ownership of products
in turn, as they flow to the end of the channels. When loss
or damage occurs or is discovered while a product is in the
possession of a channel member who owns the product, its cost
is a loss to that firm. Carriers are generally liable if the
damage can be attributed to negligence in transit. But
generally, the customer firms that unpack the product bear the
greatest loss because of the accumulated damage that occurs
throughout a distribution channel while the product is

concealed by the package.

3
Distribution damage can represent significant costs to any
firm, They include replacement merchandise, wasted resources
and profit potential, return transit costs, and overpackaging
which may result from misperceptions that insufficient

packaging was the reason for the damage.

To Crosby, the cost of quality--or the price of conformance--
is generally much less than the cost of non-conformance, or
what it costs to fix the problem (1979). There is also the
ultimate cost of losing a customer who repeatedly receives
damaged merchandise. Although difficult to measure, the cost
associated with customer dissatisfaction--from damaged
products and longer lead times to get product in. good
condition to the consumer--is lost sales. This includes "not
only the margin lost by not meeting the current sales demand,
but . . .the present value of all future contribution to
profit foregone by losing a customer" (Lambert 1975, 8). The
consumer's perception of quality affects buying decisions, and
damaged merchandise gives a buyer a very clear perception of

the quality that does not exist.

Cavinato found that, although the costs are significant, the
concept of total loss and damage as an element of the firm's
logistical system has not been developed for a variety of
reasons which are listed below:

1. It represents a burden that often can be shifted
onto other firms.

4

2. Responsibility for loss and damage in a single firm
is often fragmented.

3. Performance evaluation and other' management
information systems rarely measure it in its
entirety or even separately in individual logistics
components.

4. Management often views loss and damage as a
tolerable cost that is not worth reducing or

eliminating because the cost of such effort appears
greater than the benefit to be received (1975, 6).

If managers had a good understanding of the overall costs, in
many cases they would find that the magnitude justifies the
need for a system to control product quality during

distribution.

In order to control quality, it must first be measured. Most
companies subjectively evaluate their distribution damage.
Company personnel, warehouses, carriers and customers may all
have some:notion.of the quantity of damage that occurs, and an
estimate of its effect on sales. But these numbers can be
greatly under- or over-estimated, and are misleading. It is
very easy to see why.
"Damage can be a function of such factors as throughput,
general housekeeping, the quality and training of
management and labor, the type of product, the protective
package used, the materials handling system, the number
of times that the product is handled, and how it is
handled. To say which of these factors is most important
and how much damage each one accounts for is extremely
difficult" (Robeson 1985).
Further, although many firms do go through some type of

quality audit and corrective action after an exceptional

5
occurrence of damage, most.do not.have a formal quality system
to routinely control the quality of their products in
distribution. Even if damage numbers are available as they
are in transit claims, they may be simply accounted, with no
action to solve the problem. "As William E. Conway said,
measurements of productivity are like accident statistics.
They tell you that there is a problem, but they don't do

anything about accidents" (Deming 1982, iii).

223295:

The purpose of this research was to investigate and compare
the current management methods of five companies in one
industry for controlling product quality in distribution. The
firms' procedures were analyzed in case studies. Each company
was examined for how their existing methods conform to the
formal quality control model offered in Chapter Two. Such
damage management systems were expected to involve setting
standards and managing conformance through appraisal and
corrective action, including package or handling improvements.

This is also discussed in Chapter Two.

In addition, this A thesis yields observations and
recommendations for managing damage information, which
involves collecting and analyzing information on a continual
basis. It also seeks to determine the potential benefits of

having a comprehensive distribution damage information system.

Magus

Packaging professionals often occupy the unique position of
being indirectly responsible for maintaining the quality of
goods as they change ownership and location throughout a
marketing system. Packaging is usually the focus of blame for
damage in distribution (even though a packaging solution to a
damage problem may be the most costly one). The first
question that a Packaging expert asks, when confronted by an
alleged "damage problem" is, "how much?" and "what kind?"
Such questions are the source for realizing a need for a

damage information system.

QLLBIEIQATIQEE
mm

It is important to note that managing damage information is
different than filing transit claims, although the former may
include the latter. Managing information about damage that
has occurred in the past goes beyond filing claims. (A company
may do both: someone may analyze the damage and decide to
change the packaging, while simultaneously a claim may be
filed for the damaged shipment. But they are two separate
issues. Filing a claim is a legal response to damage that has
already occurred, while analyzing causes and types of damage
is done to prevent it from occurring again. In addition,
claims are only filed for in-transit damage by the party that

owns the goods during shipment, and they may not show the

whole picture:
Claim form analysis is one source of damage information.
The number of units and monetary loss can be found in a
claim filing. Not caught in this approach is concealed
loss and damage that is not detected until after a time
period has passed in which a claim can be filed as well
as small claims that the firm does not file (Cavinato
1975, 43).

On the other hand, damage management information systems can

span the boundaries of the firms in distribution channels who

own, handle, and transport a product.

£122g_2f_nsnsgg
This thesis did not attempt to compare and contrast types of

damage or different types of marketing systems, although they
may appear in discussion. The intention was to compare how
some similar firms manage infigzmatign about damage that has

occurred.

L2£§_BR§_D§E£§2

Often the terms "loss" and "damage" are used together. The
research in this thesis was limited to products whose value
has been reduced by damage at some point in the distribution
channel. It did not include merchandise that has disappeared
because of theft, counting errors, and other causes of product

loss.

CHAPTER II

LITERATURE REVIEW

The abundant literature on quality control addresses ways to
improve product quality and, reliability. Also common are
methods for reducing the costs of controlling quality, either
on an organization-wide basis or at a very detailed level.
This literature review begins by exploring the philosophy of
Total Quality Control. It then discusses the well developed
quality concepts pertinent to production and manufacturing,
and finally, explores the minimal literature related to damage

in distribution.

IQIAL_QQALIIX_QQNIBQL
In the opening chapter of IQ§§1_anlity_§Qntrgl, Feigenbaum

states that the goal of competitive industry "is to provide a
product and service into which quality is designed, built,
marketed, and.maintained at the most economical costs," while
still allowing for full satisfaction of the consumer (1983,
5). This statement is based on the belief that control of
quality "must start with identification of customer quality
requirements and end only when the product has been placed in

the hands of a customer'who remains satisfied" (1983, 11). To

9
achieve this goal, the actions of people, machines, and
information must be coordinated through total quality control.
To Feigenbaum,
Total quality control is an effective system for
integrating the quality-development , qual ity-maintenance ,
and quality-improvement efforts of the various groups in
an organization; so as to enable marketing, engineering,

production, and service at the most economical levels
which allow for full customer satisfaction (1983, 6).

Juran describes how traditional planning for quality, which
was delegated to the functional departments, failed to
optimize company performance relative to quality. He calls
today's process of strategic quality planning, or managing
quality on an organizational level "Companywide Quality
Management - CWQM" (1988, 6.23). Similar to Feigenbaum's
total quality control, this concept begins with broad
corporate quality goals and then objectives are "deployed"
into specific responsibilities and procedures. He describes
quality control as "the regulatory process through which we
measure actual quality performance, compare it with quality

goals, and act on the difference" (1988, 6.31).

Sinha.and Willborn use the phrase "Total Quality Assurance" to
suggest that:

. . .total quality control does not usually mean a
cradle-to-grave system with regard only to the product
life. [It requires] the total involvement of all staff
in an organization together with suppliers, distributors,
and even customers, in bringing about quality and
satisfaction. . .. (1985, 23).

10

Total quality embraces all functions and activities of the
organization. At the uppermost level, it is a policy or a
position, and it may be stated in a very broad manner. For
example:

". . .to provide a product to our customers which

satisfies their performance, quality, reliability, and

safety requirements at a fair market price. . .."
Such a goal is oriented to the results the organization would
like to achieve. It is not specific as to the means necessary
to accomplish goals. Hence, "the deployment process consists

of allocating the goals to lower levels of the organization"

(Juran 1988, 8.6).

TEE_EQDEL

Many authors propose models to describe the quality deployment
process. In "Quality and Control in Logistics: A Process
Model," Robert Novak did an extensive literature review of
existing control models to form a comprehensive model of his
own for controlling logistics functions (1989). Based on his
and this author's research, the model shown in Figure 1 will

serve as the basis for discussion in this thesis.

Novak finds that there are three basic steps in the models he
'reviewed:

(1) the development of goals, standards or plans

(2) performance measurement or variance analysis

(3) feedback or corrective action.

11

Hire 1: Quality Model

monh‘or
process

1 3
5e? analgze
standards ,conformance

4

planned correc iue

improvemen’rs acfion
long-farm focus men? Fumbles

 

 

12
This model involves setting standards; and then managing
conformance to these standards by collecting and analyzing
information, and taking corrective actions. This is a closed-
loop system, because the process should be continuous.
Corrective actions and planned improvements can lead to the

setting of new standards.

Functional levels can utilize this system to meet their
individual quality goals. Because there are so many functions
in a firm, there is a wide array of standards and methods for
monitoring conformance due to the diverse goals that need to
be met. This will be discussed in more detail later in this

chapter.

Most authors in the literature reviewed discuss breaking down
broad quality strategies into specific responsibilities for
functions. Juran includes supplier relations, manufacturing
planning, production, marketing and customer service, quality
assurance, and administrative and support functions (1988,
5.1). Feigenbaum lists main-line operations as marketing,
engineering, production, industrial relations, finance, and
service as well as the quality-control function itself (1983,
13). There is mention of the same, or similar general
functions by many authors. Many, like Deming, offer specific
methods for achieving quality, and their literature offers the

most deployment details in the production arena.

13

t

REQDEQIIQN_QQLLIIX
Just as corporate quality strategies attempt to guarantee the
quality of the overall service a firm offers, quality control
in production processes consists of assuring the conformance
of product and process to standards. There are many ways to
measure performance in the production environment, depending
on the process. Feigenbaum offers quite a few examples:
. . .activities for specifying engineering tolerances in
user-oriented terms, accelerated test methods for
evaluating component and systems reliability, classifying
quality characteristics, vendor rating methods, sampling-
inspection techniques, process-control techniques, design
of quality-control measuring equipment, computer-based
quality data processing, gaging systems, standards
establishment, product-quality' evaluation and. rating
- schemes, application of statistical techniques from X and

R charts to designed experiments, and many others (1983,
6).

Many authors illustrate methods to apply quality principles in
production. This section explores how each of the quality
stages from the model described earlier are deployed in

production.

at 8
Crosby defines quality as "conformance to requirements" (1979,
17) . Standards must first be set to operationalize this

definition. Only then can conformance be measured. Standards

 

*For the purposes of this thesis, the term "production"
refers to all activities involved with the actual fabrication
and manufacturing of the product.

14
in production consist of specifications and.tolerance limits,
designed to assure performance of machinery and other inputs
to the processes. Among the common quality measures used to
evaluate shop floor performance are reduction in percent
defective (because defects are a symptom of a system out of
adjustment), reduction in error rate, and reduction in
variation around a target value (Baker 1988, 10.50). These
measurements are used to adjust equipment, behaviors, or other

aspects of the system that are out of control.

Managing quality in production is much more objective than
managing total quality concepts, because once standards are
set conformance can be measured. To measure conformance, the

processes are monitored by collecting information.

We:

There are two ways to monitor a process. The old-fashioned
way is to conduct audits periodically. The other way, which
is more appropriate for quality control, is to monitor the
process on a continual, on-going basis. Production processes

often employ methods that monitor on a regular basis.

The best way to collect information is to utilize a management
information system designed to collect, analyze, and report
findings. Sinha and Willborn suggest that "the main data for

quality assurance are those acquired from the monitoring of

15

defects and failures and any procedures and other measures for
prevention.” They list the types of measurements that should
be included:

1. Kind of defect.

2. Cause for defect.

3. Location of defects detection.

4. Weight and relative significance of defect.

5. Evaluation of defect.

6. Cost of defect.
They add that "information must be relevant, reliable,
current, sufficient, understandable, and optimal in every way,

so that decisions are made in a timely, accurate and effective

manner" (1985, 479—483).

W

"Most decision making in quality control...rests on a base of
statistics--defined.narrowly as the collection, analysis, and
interpretation of data, or more broadly as 'the science of
decision making under uncertainty'" (Dudewicz 1988, 23.1).
The Imost. common. means of reaching' quality' decisions is
regulating all stages of production with statistical process
control, which consists of using the basic tools of frequency
distributions, control charts, sampling, and tally sheets
(Juran 1988, 6.21). Statistical process control and the first

two of these tools will be discussed further.

statistical Process Control (8P0), as defined in the Quality
W is "the application of statistical techniques

for measuring and analyzing the variation in processes."

16

There are two components of variation in manufacturing
processes: a steady component which is inherent in the
process--called ranggm variation; and an intermittent
component--attributed to W causes. Dr. Walter A.
Shewhart, of Bell Telephone Laboratories, developed this
theory of statistical quality control in the 1920's. "He
concluded that assignable causes could be economically
discovered and removed,. . .but that random causes could not
be economically discovered and could not be removed without
making basic changes in the process" (Shainin and Shainin,
1988, 24.3).

A repetitive process or operation will seldom produce

exactly the same quality, size, or other measure to be

controlled; rather, with each repetition the process will

generate variation around some average. Because this

variation is usually due to a large number of small,

uncontrollable sources, the pattern of variability is

often well described by a standard frequency
distribution. . ..(Meredith 1987, 526).

Frequency distributions (or histograms) are statistical tools
for "presenting numerous data in a form which makes clearer
the central tendency and the dispersion along the scale of
measurement, as well as the relative frequency of occurrence
of the various values" (Dudewicz 1988, 23.12). Frequency
histograms are often used for comparison of process

capabilities with tolerance limits.

Deming warns, however, that frequency calculations "serve no

useful purpose for improvement of a process unless the data

17
were produced in a state of statisticalcontrol." Data points
may fall within a normal distribution, but one would not
recognize a downward trend, for example, unless the points are

plotted on a control chart (1982, 114).

Control Charts aid in distinguishing between the two types of
variation, and are used to detect when a process has gone out
of control. .Shainin and Shainin state that control charts are
commonly used to:

1. Attain a state of statistical control.

2. Monitor a process.

3. Determine process capability (1988, 24.6).
Limits are set to»determine when an average of measures is too
high or low, commonly plus and minus three standard
deviations. "Lack of control is indicated by sample averages
arising when the process is likely to produce single items

outside specification" (Thomas 1965, 155).

Deming suggests that everyone in the company understand
statistical reasoning and be able to use elementary
statistics. He advocates that the following statistical
techniques should be taught to all workers and managers:

1. How to read and construct a histogram

2. How to read a process flow chart

3. How to construct an Ishekaawa ("fishbone") chart
4. How to understand a Pareto chart

5. How to read x-bar and R-bar charts

6. Scatter plots (correlating x and y). (Tribus)

18
There are many different causes for variation described by
statistical distributions and control charts. "The Egreto
pninginlg tells us that a few of these causes will have a
major effect on the total variation. A few more will have a
somewhat lesser effect. Most will have a very small effect"
(Shainin and Shainin 1988, 24.5). The Pareto breakdown can be
done at virtually any level (from the organization and
personnel to the type of defect) due to the endless variety of
sources to consider. Because the Pareto principle helps to
identify the magnitude of causes, "it helps to assure that

resources and attention are concentrated where they will do

the most good" (Juran 1988, 6.20).

 

Once the causes are found, corrective action on the current
problem can be carried out and longer-term improvements can be
planned, including the review of standards. Examples of
corrective actions in production include: changing suppliers
due to materials not being "up to spec," adjustment of
machinery, and re-training of employees who may have strayed
from procedures due to conflicting production performance

standards.

19
QEALIIX_HQDEL_L££LI£ATIQN§
Figure 2 summarizes the stages from the Quality Model and
relates them to broad corporate and more specific production
applications. Applications in distribution will be added in
the Results chapter, where this figure will appear again,

after a comparison of the cases.

 

 

 

 

 

I CORPORATE PRODUCTION DISTRIBUTION
SET Corporate goals Types of
STANDARDS Customer defects

service Levels of
Strategic acceptance
planning Product
Production specifications
rates Location of
damage
MONITOR Monthly reports Inspection
PROCESSES Customer Sampling
feedback
ANALYZE Compare reports SPC - Frequency
CONFORMANCE to goals distributions,
control charts
CORRECTIVE Project teams Review
ACTIONS Quality circles standards
& Education 8 Adjust
PLANNED training equipment
IMPROVEMENT Reevaluate Education
goals Reevaluate
suppliers

 

 

 

Figure. 2: QUALITY MODEL APPLIED TO CORPORATE AND PRODUCTION LEVELS

 

 

 

20

Q!ALII!.IN_DI§IEIBQIIQN

If broad total quality concepts are narrowed down to
production.and.other functions, it seems only natural to apply
them to products in distribution. Unfortunately, ". . .too
often we go through the whole process of supplying goods and
services, all the details of contracting, requirements
planning, testing--and yet fail to look at that final item,
distribution" (Esterby 1985). Many logistics authors discuss
quality in distribution and customer service in terms of
completeness of orders, order cycle time, etc. Literature
about cuStomer satisfaction also includes product performance
once in the hands of the consumer; However, control of damage
has generally been neglected in lists of the objectives of

distribution quality improvement.

Feigenbaum (1983) and Juran (1988), in their third and fourth
editions respectively, do not specifically discuss quality
with regard to distribution damage, although Baker recognizes
the need:

The need to measure goes beyond the "hardware"
departments involved in production: it extends to the
"software" departments performing support activities--
‘engineering, accounting, data processing, finance,
marketing, material control, etc. While all these
departments have quality problems, not all of them have
a quantified measure of quality" (1988, 10.50).

Gryna also supports this finding:

Those activities which directly influence the nature of
the product (e.g., design, purchase of materials,
fabrication, inspection) have received much attention
from the quality 'movement.' However, there are other

21

activities which, though indirectly influencing quality

of product, have largely escaped the attention of the

quality movement (1988, 21.2).
Gryna calls these functions that are generally lacking in the
quality literature "administrative and support activities."
Included in administrative activities are finance, personnel,
data processing and computers, plant engineering, legal, and
other activities. Listed as support functions are shipping,

receiving, material, in-process and finished goods storage,

traffic, and order filling, among others.

Although these functions seem to have been neglected by the
quality movement, Gryna does say that "quality control
[itself] has long been practiced in administrative and support
activities." He goes on to discuss several tools of diagnosis
and how they are identical with those used in product-oriented
quality improvement. His discussion examines ways in which
quality concepts can be applied to the administrative and
support activities, as this thesis seeks to do specifically

with distribution damage (1988, 21.2).

As in production or any other function, achieving product
quality goals in distribution involves setting standards;
managing conformance by developing a system for collecting and
analyzing distribution damage information; and taking
corrective actions, which may include package, product,

transport, or handling changes.

22

Esssin§_th1_fiisngssgg

To set standards for the control of product quality in
distribution, valid physical measures are required. Critical
defects must be defined, and threshold levels must be set for
judging acceptability limits. The standards should reflect
controllable variables: common types and causes of damage, as
well as cost associated with damage. The standards should
also "reflect.both.short run.and long run.performance and.they

should be reviewed on a periodic basis" (Novak 1989).

The standards should be set by a team which includes
representatives from the involved departments: Packaging,
Manufacturing, Engineering, Quality Control, Logistics,
Material Handling, and Marketing (as well as customers), and
any other functions affected by product damage. Since these
people will also be the quality management system users, they
will be able to tailor the standards to meet their specific

needs.

Once standards have been set, methods must be developed for
appraising the conformance of products to the standards.
Bowersox, when discussing strategic logistical leadership,
says that ". . .measurement is the regular and meaningful
assessment of performance. If you are not measuring
performance in a variety of ways, you are not managing your

business" (1990)

23

MW
Appraising conformance requires that processes be monitored,
in order to obtain information for analysis. There are many

methods to monitor performance. Some examples follow.

Mitsubishi Electric Corporation, a major Japanese manufacturer
and marketer, monitors conformance to standards by using a
standardized form. Package and product damage are reported in
detail by distribution centers and sales companies within
seven days after finding the damage. Using this form, the
causes of the damage are investigated in order to improve
products, packaging, and distribution practices (Maezawa

1987).

Field. audits are another' method of investigating' damage
complaints. Often, when long-distance feedback is the only
source of information, data is incomplete and fragmentary or
totally wrong because the people relaying the information are
not well qualified to evaluate damage. Teams of researchers
who inspect routine shipments and/or test shipments can get
first-hand information about the specifics of product damage.
Although they are valuable for guiding corrective action,
field audits are often only performed when a known problem

exists, and they do not measure the magnitude of a problem.

24
To gather both common and abnormal occurrences of damage, a
method of appraising distribution quality on a regular basis
should be employed. An information system can be used to
formally feedback details of shipments. In order to be
useful, such a system requires institutionalized detection,

measurement, feedback and analysis of damage information.

Management 91 Information. This is the age of more

information for lower cost. There is a wealth of literature
about management information systems. There is an increasing
emphasis on information systems in logistics, because better
information can reduce the cost of customer service. However,
although logistics writers discuss the importance of a timely
and accurate information system for the effective control of
logistics, they' do» not. generally' propose tailoring' such

systems to manage information about distribution damage.,

A quality management information system must begin at the
first moment of damage detection. At that moment, the most
clues to the causes of damage are available. Products may be
formally inspected at some point during distribution or the
detection may be by chance, such as when a worker handling a

product notes a problem.

At the moment that damage has been detected, it should be

measured against the standards for product/package quality.

25
For example: Is the dent larger than the size allowed by the
standard? Is the scratch critical?"Where is it? ‘Why/how did
it happen? In order for the person detecting the problem to
be able to generate this kind of data, he/she must be educated
to observe and classify damage. This education provides an
additional benefit--the sensitivity to damage issues

encourages material handling workers to be more careful.

Once damage has been detected, a method is required to feed
back the information to those who can analyze the data and
correct the causes. Transmission of data can be either
electronic, verbal by phone, or by mail. Electronic data
interchange (EDI) is the quickest method, tying damage

information to other logistical and order information.

In W. Gryna offers actions to
improve the quality and.promptness of field feedback; Some of

his recommendations should be incorporated into any system
involving feedback, including distribution damage information:

1. Provide personnel with well-designed data sheets.

2. Provide incentives to encourage adequate feedback.

3. Provide a glossary of terms to improve
communication and a mnemonic code number to
simplify the data entry and analysis.

4. Provide training in the how and why.

5. Conduct audits of the data feedback process.

6. Make use of modern technology to collect the field
information.

7. Make use of modern methods of analysis to provide
managers with valid summaries for decision making.

8. Minimize the number of data relay stations.

9. Obtain the operations log (1988, 20.28).

26
It should be noted that throughout the system, the amount,
accuracy, and speed of data collection and transmission
depends upon the type of marketing channel. Vertical
marketing systems, in which the producing company owns or has
a contractual relationship with its channel participants,
offer more control over logistical information than do non-
vertical marketing systems, in which products change ownership

more times.

MW

There are many opportunities in distribution to measure
performance by collecting and analyzing information. For
analysis, correlations can be drawn between damage types,
stock keeping units, costs, carriers, shipping routes,
distribution centers,. . .whatever information is most
relevant to making decisions about damage prevention. Reports
are prepared which rank products, damage types, carriers,
distribution centers, and costs, to illuminate the biggest

targets for improvement.

Gryna discusses some tools.of'diagnosis that could be used for
quality improvement in administrative and support functions.
These tools are identical to those used in production-oriented
quality improvement projects. Among them are:

Pareto Analysis

Cause-and-Effect Diagram

Frequency Distribution and Histogram
Dissection of a Process

27

Statistical Control Chart
Sampling (1988, 21.16).

All departments who are in.a position to correct the causes of

damage should receive reports of data and correlations.

 

If no corrective actions or planned improvements result from
reporting, collecting, and analyzing damage information, the
whole purpose of the system is lost. Corrective actions are
used to realign actual performance with standards to correct

the causes of damage.

Information tends to get less detailed and problems get
blurred as they move farther from the problem site. Because
of this, the ideal situation is to have the person closest to

the activity take corrective action.

Most importantly, corrective action should address the causes
of problems, not only the problems themselves. Like defects
on the shop floor, distribution damage is a symptom of an
underlying problem. . .a quality system out of adjustment
(Twede 1989). The elements which may be "out of adjustment"
are those which cause the damage: faulty product design,
insufficient packaging, and inappropriate methods of handling,
storage, and transportation. Until the cause of the problem

is found and corrected, damage will continue.

CHAPTER III

RESEARCH METHOD AND FIRMS INTRODUCED

This chapter is organized in the following manner: first, the
case research method is discussed: next the sample criteria

and selections are described; and last, the method of analysis

is explored.

 

Product quality in distribution is a largely unresearched .
topic in both quality and logistics literature. Because this
is.a relatively new area of theory, this research employed the
case study approach. Boulton supports the use of case
research in areas of innovation and theory development:
Case research can readily be applied to new areas which
require systems thinking. In the earliest periods of
research, long before you have developed any theory, data
must be gathered in an attempt to describe the territory
and raise basic questions about its interrelationships
and processes. . ..In fact, one might argue that

statistical techniques are seldom used to improve theory,
only to accept or reject hypotheses.

Boulton argues that large sample research in management
practices does little more than determine how many firms are
using a practice or technology. Statistical analysis cannot
adequately answer questions of "why" or "how" an innovative
process was begun, or problems that may have been experienced

28

29
along the way. Nor can it adequately explore the most
advanced concepts and technologies employed by only a few

firms (1985, 3-14).

Case research in marketing theory building is explored by
Bonoma, who recommends its use when a researcher seeks to
relate phenomena in natural settings.

If properly conducted, research by these methods can
provide a "deep understanding (Geertz 1973), a fuller
contextual sense of the phenomena under study (Miles
1979) , and an explicit provocation toward theory building
that often is missing from both simple descriptive work
and most cause-and-effect research (van Maanen 1982)"
(Bonoma 1985, 201-2).

Furthermore, Bonoma distinguishes between statistical methods
of theory disconfirmation and case research in terms of the
researcher's goals:

First, the goal of data collection in case research is
not quantification or enumeration, but rather (1)
description, (2) classification (typology development),
(3) theory development, and (4) limited theory testing.
In a word, the goal is understanding. Second, most
enumeration is of little value to a case researcher. The
goal is not the breadth or representativeness of large-n
research, but rather the depth of the knowing. The risks
of low data integrity are traded for the currency and
contextual richness of what is learned (Bonoma 1985,
206).

Case research can yield a rich source of observations. These
observations lead to new hypotheses and research questions for
further investigation. "Although the case research approach

does not permit the rigorous testing of hypotheses because of

30
the limited sample, the investigation does take place within
the domain of a conceptual model" (Twede 1988, 49). This
research contrasts some companies' distribution damage control

efforts in comparison to the model outlined in Chapter 2.

IEB_£A§E_§AH2L£§

The five firms chosen are all in the office furniture
industry, a convenience sample since the companies are located
within close proximity to the researcher. This allowed for
personal interviews, rather than phone contacts only. In each
case, the key informants were identified by first
communicating with a known contact. This led to the mailing
of a proposal outlining the purpose of the research, the data
requirements including some sample questions, the potential
benefits to the firm, and a statement assuring confidentiality

regarding the identity of firm and individual names.

The initial phase was followed by a visit to the company
during which the questions were asked. Included in the
discussions were members from.departments who ideally had the
best grasp of damage information. Someone from the Packaging
department was interviewed at four firms. Other persons

interviewed were from the Traffic, Quality, and Claims areas.

The five firms are identified by letters. The firms are not

discussed in any particular order, and letters. do not

31
designate any affiliation with company names or any other of

the firms' characteristics.

TE£_H£IEQD_QI_ANBL!§IE

The analysis of responses from the case study participants was
used to deduce generalizations about current methods of
monitoring product quality in distribution and functional
benefits which are commonly considered. Boulton suggests
systematically organizing data to carry out a comparative
analysis of data between multiple case studies. "The analysis
of similar data over several organizations allows the
researcher to identify similarities and differences which
leads to the development of new concepts, language and theory"
(Boulton, 12-13). This research will follow Boulton's

approach.

The following chapter compares and contrasts the case results,
with regard to how they adapt to the model offered in Chapter

Two. Chapter five follows with conclusions.

CHAPTER IV

RESULTS

This chapter compares and contrasts the five cases. The
firms' approaches to distribution damage are compared to the
stages of the Quality Model in order to determine if and how
the firms operationalize each stage. Each section (relating
to each stage in the process) begins with a brief discussion
and states how many firms operationalize that stage. Then,
examples of individual firms' practices are given, along with
limitations of particular methods. The chapter ends with a

discussion of other related factors.

SETTIN§_§IANDAED§
Two aspects were considered regarding standardizing defects:

the level of detail, and the threshold level.

The case firms varied widely on the level of detail contained
in their defect standardization. Firm B's "poor packaging"
and "damaged in transit" are very broad and ineffective

descriptions compared to firm A's "dented" and "broken."

More detail to guide corrective action could be obtained by

using terms that refer to location on the product and the

32

33
point in the channel that the damage occurred. None of the

firms had this much detail in their standards.

Another consideration is the threshold level. Is a scratch
considered a nonconformance to standards (damage) when an
installer can fix it on site? Or is it a nonconformance only
if the scratch is bad enough to necessitate a replacement
order? In the former situation, the producing firm may never
learn of the damage, and therefore cannot take corrective
action to prevent it from occurring on a more widespread

level.

One firm sets standards: Firm A. Critical defects are
specified and differentiated by firm A to reflect common types
of damage. The standards are listed on Field Problem Reports
(FPR). FPR's are used for damage communication, shortages,
overages, and other delivery and installation problems, and
also serve as a replacement order. They are filed by
employees of firm A, who oversee problems both at dealerships
and at installations, and by customer service representatives

when dealers call in with problems.

Codes are used to represent reasons for the report. The
"where," "what," and "why" of thejproblem‘are included, as are
types of damage or "nonconformances." The following is a

partial listing of firm A's "nonconformances":

34

Bent

Bubbled
Closes/opens improperly
Cracked

Gap

Loose
Noisy/squeaky
Scratched -
Surface flaw
Torn

Warped

Broken

Chipped

Color faded/spots
Dents

Grease/oil

Misaligned

Protruding staples/screws
Separation

Too few/many
Unstable/wobbly
Wrong color/part/size

However, even if the code for "scratched" is indicated in the
column for nonconformance, that is usually the extent of the
problem description. Unless the persons passing on the
information, and filling out the order, indicate exactly what
may have caused.the damage, the all-important "how" may not be

known.

There are limitations ’to firm A's standards. Threshold levels
are not set for judging acceptability limits. Since the
seriousness of the defect is not defined, opinions may be a
factor in deciding to report damage information. There are no
standards for causes associated with damage, or costs related
directly to the damage (other than the cost of the product).
In addition, FPRs may flag' problems at dealerships and
installations, but not on the production or packaging lines,
or at loading docks. Finally, damage standards are only
applied to these reports. If one is not filed, damage is not
compared to standards, because it might.not.be known that they

exist.

35
Firms B, C, and E also have field problem/replacement order
systems. (Firm D does not). Although they may break the
"where" down into areas such as manufacturing, shipping,
installing, etc., the most detail allowed for is, for example,
"damaged in 'transit." Therefore, they’ are not setting
standards. The only way of knowing the kind of damage or how
and why it happened is if the people ordering and filling out

the form note this information in a comments section.

For firm A, having more detailed information means that they
are better able to make corrective decisions, and they are
able to do it more quickly. After exceptional occurrences of
damage, less time is spent obtaining information that could
very easily have been noted in the first place. There is not
enough detail on the other ~ three firms' forms to be of

assistance in tracking problems back to their causes.

Firm E is currently revising its replacement order system to
include codes on.types.of'damage and.other'details (i.e., they
are going to set standards). As of this writing, the only
distinguishing codes are between "noted" and "concealed"
damage. It appears that firm E's new system will be similar

to firm A's.

It should be noted that other types of standards may exist.

For example, a firm may have a stated (but not managed)

36
standard of making-~and delivering--qua1ity products. If
someone in the logistical system notices that a particular
product is damaged (i.e., it is not "up to standard"), they
may pass this information on to those who can look into the
problem. Standards such as these are extremely informal, and
they are not very helpful for solving systemic problems, due

to the number of ways they can be interpreted.

MQNIIQBIN§_T§§_BEQQE§§

As discussed in Chapter II, appraising conformance requires
that the processes be monitored in order to obtain information
for analysis. All five of the firms monitor their products'
quality during distribution in some ways. Replacement orders,
audits, and EDI are some of the sources from which the firms
get damage information. These and other methods are discussed
below, including a listing of the firms that operationalize

each particular practice.

It should be noted that the firms do not necessarily
differentiate between monitoring and analyzing conformance.
It is possible that the same method can be used both to
monitor progress, and to analyze specific problem situations.
Therefore, the same practice may appear in discussion in both

of these two sections.

37

WW

Four firms (A, B, C, E) get a good deal of their damage
information from replacement orders, which are also called
field problem reports. These reports (discussed briefly under
SETTING STANDARDS) are filed by dealers and/or customers when
a product needs to be replaced due to damage or a number of
other reasons. Many times workers will find damage during
installation, where it was concealed at the dealer's warehouse
or coming off the truck. If is is necessary to fill out a
replacement order for damaged product, it is a good

opportunity to obtain damage information.

Unfortunately, replacement orders are not always filed when
damage occurs. Dealers can replace from stock, make repairs,
etc. , and the producing firm might never know about the
damage. In addition, even if a replacement order is filed
because of damaged merchandise, it seldom indicates the true
cause in a standardized format, as indicated earlier. There

is simply not enough detail.

manila:

Audits can be used as a way of continually monitoring
distribution procedures, and they can.be useful in evaluating
specific problems. They can be conducted on production areas,
distribution centers, carriers, dealerships, and installation

sites. All of the firms conduct damage audits some times.

38

Firm A continually monitors loading at the dock, and they
conduct formal weekly audits. The auditors check for proper
loading, and other routine procedures. Firm A also provides
check-off lists with.every bill of lading, whiCh serve to flag
delivery problems on a regular basis. In addition, they have
random field audits conducted by quality assurance once a
month, where a team evaluates unloading and installation

procedures.

Firms D and E conduct audits on a random basis. Like firm A,
the- auditors observe packing, loading, and delivery
procedures. Firms D and E also have planned audits for
shipments and installations of new products, in which they
look for circumstances that may require special procedures.
Firm C sends an audit team to major installations (over a
certain dollar amount) to get a general feeling on whether
workers are educated on all aspects of unloading and assembly

or not.

All of the firms send audit teams out when checking into a
particular problem. This latter way of utilizing audits to
obtain further information about a known problem is really a
method of analyzing conformance, and it will be discussed

later.

39
WW
All five firms have electronic data interchange (EDI) or some
type of electronic communication with some dealers. Firm C
has major dealers on-line with EDI. By utilizing this real—
time customer service system, dealers can place, check the
status of, and change orders. The system is evolving so

dealers can also include information about damage.

Firms B, D and E also have electronic communication with major
dealers. They too are currently investigating methods of
utilizing EDI to measure delivery performance. Currently
there is no particular format for transmitting damage
information (this information is primarily reported in
replacement orders), but it is sometimes included in comments

or by memo.

Firm A is investigating the implementation of EDI. ‘The firm
representative said that it will enable them to have real time
visibility to damage and order information across the network.
At the present time, only firm A's customer service
representatives have access to this information initially.
They then pass it on to the quality group, who analyzes and
distributes it accordingly.

40
There are not many limitations with using EDI. For smaller
dealers (and suppliers) cost could be a limiting factor.
Another constraint could be not using EDI to its fullest
potential. If the capabilities are recognized and utilized,
EDI can improve, and reduce the cost of, customer service.

This is discussed further in Chapter five.

W

Firm A has a form, called a Corrective Action Report, which
can communicate damage information. Employees from any area
within the company fill out these forms to identify
significant issues that need correction. The system can be
used with many problems that exist; it was not designed
specifically for damage problems, although it has the
potential to identify them. Unfortunately, there are not many
of these reports filed by people in the field (the ones who
see the damage) because it is extra paperwork. .Additionally,
the forms are accessible to company employees only, and are
not made available to dealers, carriers, etc. Firm A's

Corrective Action Report program is about a year old.

Firms D and E have an internal document called a Discrepant
Material Report (DMR) which can provide damage information.

This form is filled out when damage is noticed in stock
situations, or when problems are experienced when shipping on

the private fleet. Similarly, firm 8 has a Return Material

41
Authorization (RMA) that dealers can use to return damaged

product for credit.

At firm E the DMR indicates the product, the location, and
what the problem is in general, but not much more (similar to
the replacement orders). The report often tells there was
concealed damage, but not what kind or how it happened. For
firm D, more detail is requested from private fleet drivers
who fill out the forms. They indicate the type of damage,
where it is on the product, etc. However, this detail (types
of nonconformances) is not based on available standards with
which to compare damage. It is included as comments from the

driver.

EHIXQIQ

Some of the firms survey dealers and other customers to
collect information for monitoring purposes. Some use it to
analyze conformance when confronted with a particular problem.

Therefore, surveying is examined under both categories.

Three firms collect information (monitor) by using surveys,
either written or verbal. Firm B attempts to monitor damage
information by attaching postcards to each order. By using
these short surveys, the dealer or customer can rate several
different areas, including the condition of the

packages/products upon arrival. The postcards are returned to

42
shipping, and forwarded on to packaging if there are any poor

ratings in damage, packaging, etc.

Like firm B, firm A sends service questionnaires with each
order’ requesting information. about. the. condition of ‘the
product. Not a lot of detail is requested, except for a
rating from "excellent" to "poor," and an area for comments.
Therefore, although some damage information may be discovered

through this medium, it is very vague.

Each month Firm A conducts phone surveys with warehouse
supervisors at dealerships to obtain information on damage,
packaging, and related'quality problems. The interviewer asks
questions following a written form, and notes areas that need

attention. The surveys are then compiled for analysis.

Firm D does not send out questionnaires on a regular basis,
but conducts surveys when new product lines are introduced and
major packaging changes have been.made to measure the success
of the innovation or change. Firm B also does phone surveys

periodically to get information on packaging problems.

Surveys of any kind have limitations placed on them by those
being surveyed. Monitoring information by survey is
respondent driven. . .information is relayed only if they take

the time to respond. The firm B representative estimated that

43

they only receive ten postcards at the most each month in

response .

In addition, information obtained from surveys is very
subjective due to a lack of standards. Someone might circle
"poor packaging," but might not state why, what particular
product was damaged, or how it was damaged. Even if the
product and type of damage is indicated, it may be one out of
thousands of shipments“ For these reasons, it is difficult to

know the magnitude of the problems.

W

All five of the firms use anecdotal damage information.
However, because it is not solicited either on a periodic or
an on-going basis, it is an extremely informal method of
monitoring. Each of the firms, however, consider this type of
information valuabLe. One of the case interviewees stated
"the company still depends a fair amount on informal
communication, and it is the best kind of information
sometimes, but you do not get it until it is already a big

problem."

Word-of-mouth information can come from dealers and customers,

company field representatives, members of industry

44
organizations, carriers, and anyone that interfaces with the

product.

The limitations of verbal communication are the delayed timing
and lack of measurement. Often, information is not received
. until a big problem already exists. Any estimate of damage
magnitude depends on the skill of the storyteller. A dealer
may call and say that.a particular type of damage happens "all
the time," when in actuality it is a very small percentage of
the time. Therefore, it can be difficult to quantify verbal
information to get a real-time measure of the magnitude of a

problem.

Another limitation of word-of-mouth communication is that
damage information may pass through several people (and their
perceptions) before it reaches someone that can use it. By
the time it gets there, the information may be ambiguous or
misleading. For example, one of the representatives said they
have trouble getting accurate information from some dealers'
warehouses. The workers' personal opinions and attitudes have
a bearing on what, and when damage problems get passed on. If
information is not communicated immediately, it may be very
vague by the time someone at the dealership, and in turn

someone at the manufacturing firm, is informed.

45

9.131.131

Generally, none of the case firms process claims for in-
transit damage. Except under specific circumstances, the
dealer or customer is responsible for filing freight claims
with the carrier. However, the manufacturer might get some
information about damage from carriers that have had claims
filed against them. Information availability varies between

carriers, and may depend on the shipper-carrier relationship.

Firms B and D request monthly reports from many of their
carriers. Firm D's reports list the amount and status of each
claim filed that month, and there are columns for product and
damage information. Unfortunately they are usually not
detailed enough to suggest corrective action. Carriers may
collect information concerning what products were damaged and
what happened in general (including shipment-specific details
like which terminals, switching yards, or vehicle equipment
were used), but the actual type and cause of damage are often
missing. Another limitation is that the carriers can pass on
information about in-transit damage, but not problems in other

areas of the logistical system.

Firm A conducts a carrier phone survey each month to get,
general damage information and opinions on packaging,

procedures, etc.

46

ANALXZIN§_£QNIQBEANQE

As stated in Chapter Two, most decision making in quality
control is based on statistics. Since the case firms receive
fragmented damage information, statistical analysis techniques
are not possible because distribution processes are not
‘monitored. on. a continual basis“ Hence, .infignngtign__i§
WWW once problems have

already been identified.

All five firms analyze conformance informally. Three
practices were discovered: quality reports, audits and
surveys.

MW

One way of analyzing product/package conformance to standards
is to prepare reports which rank products, damage types,
carriers, distribution centers, and costs, to illuminate the
biggest targets for improvement. Correlations can be drawn
between the above mentioned, stock keeping units and shipping
routes,. . .whatever information is most relevant to making
decisions about damage prevention. This process is similar to
Pareto Analysis, although Pareto requires formal data
gathering to be effective. For the case firms, the only data
available for analysis is that which someone has decided to
convey. Most of the firms' reports are prepared by their

Quality departments.

47

Firms C and E prepare corporate quality reviews on a quarterly
basis. They combine quality issues and highlight trends and
particular problems using information gathered from the damage
sources listed in MONITORING CONFORMANCE. For firm C, the
reviews concentrate on product failure and damaged freight,
and they are based on information that is obtained by asking
specific questions during a reorder. Firm E's reviews cover
all quality issues. Because of this broad scope, other
problems (for example, production defects) may outweigh
distribution damage each quarter. Both firms C and E use the
reports to focus on major problems, and they assign resources

and project teams to further analysis.

Similarly, firm D prepares monthly reports based on
information received from customer and dealer complaints,
replacement orders, and DMRs. The damage reports, which
highlight the products that suffered the most damage that
month (based on the best information available), trigger

investigations into problems.

Firm A uses their surveys (discussed earlier) to compile
monthly reports. Then, correlations are drawn to better
pinpoint damage. For example, by using the monthly reports,
they could determine that some of their east coast.dealers had

experienced a marked increase in damage to file cabinets that

48
month. They could then investigate their carriers delivering

to the East coast to find out why.

In addition, Firm A can print reports that summarize the
Corrective Action Reports, or the Field Problem Reports
(replacement orders). The data base can be sorted by any of

the categories contained on the forms.

W

As mentioned in monitoring conformance, audits and surveys are
both a common means of finding out about problems, and of
quantifying particular problems that the firm is already aware

of. All of the cases said they use both practices to further

analyze problems they already know about.

 

Corrective actions should address the causes of problems, not
only the problems themselves. For example, damage can result
from rough handling during production, handling, transport or
storage procedures. But behind that rough handling may be a
lack of management, assuring that things are being done

correctly.

All five of the firms take corrective actions and plan for
improvements. Some of the corrective actions that the firms

have taken include forming project teams, making packaging and

49
product changes, and training employees. These and other

actions follow, and they include specific examples.

W

All of the firms form special project teams for solving some
problems. For example, firm A acts upon the corrective action
reports discussed under W by having a
committee appoint the necessary people and resources for
follow up. The follow up action taken is formalized, and
there are standard operating procedures to follow. In a
damage situation, teams evaluate various aspects of the
distribution environment, asking specific questions about the
problem being investigated to find the cause. Some of the

corrective actions implemented by project teams follow.

W

Packaging changes are often made to solve a damage problem.
A solution might be to use stronger materials, change the
package altogether, or eliminate the package and blanket-wrap
the product (as is commonly done in the office furniture
industry). Firm D pointed out that eliminating the package is
not a solution in some situations, because customers and
dealers can choose to have all products in packages. However,

they encourage the use of blanket-wrap.

50
Firm C had a problem with scratches on the bases of their
chairs. .Although the scratch could not be seen when the chair
'was assembled and in place in an office, it could be seen when
the carton was unpacked because in the double-pack box, one
bottom was facing up. A Japanese distributer for firm C's
product considers the nonconformance unacceptable, and notes
this on monthly status reports. The solution was to use a
returnable rack package to hold the bases, which are

individually wrapped in a kraft paper to avoid scratching.

Due to a lack of information about handling capabilities,
procedures, etc., it is often assumed that the package is at
fault and a change is made, even though that may not be the
best solution. This often results in a very expensive package
that protects the product no better than the original one did.
Additional cost per package is a direct variable expense, and
may also increase the cost of solid waste disposal, whereas
management corrective actions, like training employees in
handling procedures, returns an investment. In addition, due
to the lack of an information system, the performance of the

new package cannot be measured accurately.

W
Investigation may reveal that the cause of the damage is due
to the inherent fragility of the product. Engineering changes

may be one of the corrective actions taken, to make the

51
product more durable for the distribution environment. In one
example, reducing the number of parts in a light fixture not
only decreased costs, but reduced the number of parts

susceptible to damage.

WW

An example of establishing a new procedure is adding an
additional inspection point to flag potential problems before
they occur, before packaging for instance. The ease of
establishing new procedures depends on the severity of the
change. ' Some procedural changes may be very simple to
implement. On the other hand, any new procedures can be
difficult to establish without the proper planning, education

and implementation.

Some of the respondents said that management is often
resistant to changing procedures until there is enough
documented justification to do so. This documentation is hard
to obtain, without first trying the new procedure. Therefore,
procedural changes can be difficult corrective actions to
take. There can be a massive number of people and steps
involved, and it is difficult to get people to change how they
do things if they have been doing them one way for a long
period of time. It is important to inform employees why

things must change, and how the change will be beneficial.

52

WW

Having the same reoccurring problems could mean that standards
need to be reevaluated. It is possible that they are too
stringent, or too lenient, or that the situation has changed
enough to dictate a reevaluation. It also is very possible
that workers are operating under conflicting standards. In
one example, a particular corrugated box is to be closed with
two pieces of tape on each side. However, due to a competing
production standard that requires the worker to handle a
specific number of pieces per hour, he only puts one piece of
tape on each side. This can lead to the box opening in
distribution, lending itself to damage. I To correct the

situation, one or both of the standards must be reevaluated.

v u d i
It is possible that damage is due to inputs into the product
or package--inventory or materials that do not meet
specification. This could affect product/package performance.
Investigation might reveal that a supplier is having problems.
Corrective actions could be taken to help the supplier
overcome any limitations, material specifications could be

reevaluated, or a new supplier could be found.

Two of the firm representatives said that they have had
problems with materials not meeting specification, or that

they have had to set standards for material performance. Like

53
the example above, where the worker was only using half the
amount of tape, the wrong kind of tape can also be a problem.
At one firm, the purchasing department found that it could
save money buying cheaper tape. They did not realize that
this new, less expensive tape was leading to higher damage
costs because it had a poor adhesive. Corrective action
involved educating purchasing on the necessary qualities of

the tape to be used.

Wigs

Training and education are valuable actions to take in all
situations. The more that people are aware of how their work
affects other aspects of the company, they will realize why
procedures exist. Videos are a common method used to further
the education of employees, showing correct procedures and
instructing workers on specific tasks. Unfortunately, videos

(and training in general) can be time consuming and costly.

Firm C has found that over time, workers stray from standard
procedures to take short cuts (again, possibly due to
competing goals) . They have recently completed making a video
to remind workers of a particular packaging process. In this
case, a corrugated slip sheet was to be laid on the correct
size skid so fabric-covered panels could be stacked on top.
However, workers would use any skid, even if it was too small,

and panels were getting scratched from hanging over the edge.

54
Also, they were omitting the corrugated slip sheet, and the
bottom panel's fabric was being torn from nails on the skid.
Hopefully, reminding the workers of standards, and why they

exist, will solve the problem.

men
Another corrective action taken is the purchase of equipment.
This rangeS' from buying new hand carts to financing
sophisticated loading and.unloading equipment. New purchases

can prove costly, but justification can often be feund in

improved efficiency.

 

Products change ownership and liability so often in logistical
systems that large totals of damage can go unnoticed because
single logistical activities may experience small amounts of
damage, or what someone feels is "insignificant." One of the
case representatives said "there really hasn't been much of a
damage problem," though he had no totals for system-wide
damage to back up his statement. If total damage is, for
example, only two percent of $100 million in sales, 2 million

dollars is being taken from profits!

Dealerships can sometimes replace merchandise through their

own warehouses. The manufacturer might not learn about these

55
replaced (damaged) products until a periodic inventory count
or an audit is conducted. There is a lost opportunity when
inventory is taken monthly or quarterly because damaged
merchandise is lumped into one category, without detailed

reports.

Trained workers can make many cosmetic changes and small
repairs on office furniture at an installation. Often, these
types of problems exist for a long time before they are
communicated back to the manufacturer. For example, by trying
to cut down on the number of employees needed for an
installation, one firm discovered that much of workers' time
was spent repairing a scratch that appeared on the same
product every time it was shipped. After years of having the
same problem occur, the firmwwas able to look into the problem

and take corrective action to avoid the scratch in future

shipments.

 

All of the firms agreed that more damage seems to occur with
less-than-truckload (LTL) shipments than with truckload (TL)

carriers (although they had no data to confirm this).

Shipping by truckload offers the opportunity to package
differently than is possible with LTleue to the fewer number

of times the product is handled and the additional control

56
over stowage and shipping conditions. For example, office
furniture is shipped in conventional corrugated fiberboard
shipping containers by LTL carriers, and is often shipped
uncartoned and blanket-wrapped by full truckload. This
results in lower packaging material purchasing and solid waste
costs. Each of the firm representatives mentioned a desire to

have more TL shipments.

d a d o ed r
All of the key informants voiced a concern about the lack of
knowledge about dealers' unloading capabilities. Often a
shipment will arrive, and the proper unloading equipment is

not available.

Firm A gave this example: in New York City many customers and
dealers lack a loading dock. Workers must unload the truck at
street level and sometimes carry product up several flights of
stairs. Since workers are often on a time schedule for
deliveries, they unload as quickly as possible. Hence, if
there are only two workers, and one is busy bringing freight
into the building, the worker outside often drops the package
from the back of the truck to the ground (a four to five foot

drop).

57
The firms also expressed concern over material handling
workers not being trained in unloading procedures. Warehouse
employees at dealerships are often part-time help, not trained
due to the short duration of their employment.(for example, a
student with a summer job). Problems also surface with
products that need to be handled in a special way. Even
trained employees may not be informed of a particularly

fragile or awkward product's handling procedures.

A 0 AP
Figure 2 can now be completed. Actual applications discovered

from the case firms are included in the Distribution column.

58

 

 

 

 

 

 

 

 

 

 

CORPORATE PRODUCTION DISTRIBUTION
SET Corporate goals Types of Damage types
STANDARDS Customer defects Level of
service Levels of importance
Strategic acceptance Causes of
planning Product damage
Production specifications
rates Location of
damage
MONITOR Monthly reports Inspection Replacement
PROCESSES Customer Sampling orders
feedback Audits/Surveys
EDI
Company
documents
Verbal
communication
Carrier claims
and reports
ANALYZE Compare reports SPC - Frequency Reports
CONFORMANCE to goals distributions, Audits
control charts
CORRECTIVE Project teams Review Project teams
ACTIONS Quality circles standards Review
& Education & Adjust standards
PLANNED training equipment Purchase
IMPROVEMENT Reevaluate Education equipment
goals Reevaluate Education &
~ suppliers training
Reevaluate
suppliers
Package
improvements
Product
improvements

 

New procedures

 

 

Figure 3: QUALITY MODEL APPLIED TO CORPORATE, PRODUCTION S:
DISTRIBUTION LEVELS

CHAPTER V

PROBLEMS AND RECOMMENDATIONS

The purpose of this research was to examine.how some companies
control the quality of their products in distribution. It was
expected that management methods would involve setting
standards, and managing conformance through appraisal and
corrective action. For the most part, this was found to be
true:

One firm sets standards; and all five firms monitor and

analyze conformance, and take corrective actions.

However, though all of the firms participate in some stages,
ge,‘ e l-gg 12‘7‘ ;_ e1": _ - ‘1'.- ' .‘ee e: g e e! o lo
W. The firms' methods are
reactive to damage occurrences. They are not planning for

quality in distribution like they plan for it in production.

In this chapter, problems are discussed and recommendations
are offered. The primary problem is that quality is not
controlled on a formalized basis. Because of (or due to)
this, the firms only obtain fragments of the necessary
systemwide damage information, and no one has an official

responsibility for managing product quality in distribution.

59

60
Problem 1: NO FORMAL SYSTEMS EXIST
Although the firms might formalize individual stages, they do
not have formal quality systems. They are not consciously
controlling their products' quality in distribution as it is

controlled in production.

The case firms all have important pieces of a quality system
for controlling distribution damage. 'Some of these pieces are
formalized. For example, firm A sets standards to identify
types of nonconformances on the product. However, they do not
formally operationalize the next stage in the quality system,
monitoring. If the form on which the standards are listed is
not observed (which is very possible since it is not always
filled out when damage occurs), there is nothing to compare

actual performance to.

Therefore, it is not beneficial to set standards if
conformance to them is not monitored systematically.
Likewise, corrective actions are very difficult to take
without possessing information that is based on comparisons of
actual product/package performance to standards. Taking part
in one stage is fruitless without also operationalizing the

other three.

In addition, if the system is not formalized, its potential

value is not realized. Many people within the firm and

61
throughout the channel could possess valuable information that
they are not transmitting simply because they are not aware
that they should, 'Unless all channel members and carriers are
aware of a system (or even stages of a system), and educated
to utilize it, information will not be conveyed on.a timely or

consistent basis, and may not be conveyed at all.

Problem 2: INFORMATION IS PRAGMENTED

Defect data collection on distribution activities at the case
firms is done the "old fashioned way," the way it used to be
done in production. Then, problems were handled only after
they occurred, and the information obtained was not complete
enough to make sound corrective action decisions. Today,
production processes utilize quality control to continually
appraise conformance, and to prevent problems from happening.

Distribution processes do not.

Although the case firms have many ways of lobtaining
information, they are not systematically monitoring their
products' quality in distribution. Therefore, they do not
have a complete understanding of systemwide damage. Even
firms that have many ways of monitoring do not have enough
standardized detail to make decisions without taking extra
steps to find out the necessary information. They can
identify problems, but can not quantify them. If distribution
practices followed the lead of those in production, they would

62
have a view of the total picture. Information--and decisions-

-would be much more objective than they are now.

Relayed information is fragmented, slow in being transmitted,
and not detailed enough. This is because the responsibility
for damage is fragmented throughout the company and the

distribution channel.

Problem 3: NO ONE IS RESPONSIBLE
At the case firms, no one is officially responsible for
managing product quality in distribution. Damage problems are

shifted between members of the firm and the channel.

As mentioned in Chapter One, Packaging Professionals often end
up being held responsible for solving damage problems since
the package is usually the focus of blame for’ damage.
However, because the system is not formalized and information
is fragmented, Packaging experts do not get the necessary,
detailed answers to the questions of "how much?" and "what

kind?" (of damage). Unfortunately, neither does anyone else.

63

The firms should take the stages that they already have
partially implemented, and develop a damage management system.
Firm A operationalizes each of the four stages from the
Quality model. They now need to formalize the stages into a
system in which all channel members participate. Firms can
follow the Quality model set forth in this thesis. It can be
used for formalizing methods of setting standards, monitoring
the processes, analyzing conformance and taking corrective

actions/planning for improvements.

Ethan Allen has a benchmark formalized system for managing the
quality of their furniture throughout distribution. A team
sets the standards for product quality (eg. breaks, cracks,
upholstery defects, warpage, crushed, missing items, finish
defects, etc.), critical locations of damage on the product
and in distribution channels, package condition, and handling
problems. Product is monitored throughout the distribution
channels, and damage information is transmitted by phone and
keyed into a database. Products are then ranked by damage
frequency as well as by highest cost losses, and corrective

actions are taken (Twede).

64
Formalizing a system will engage the participation of the many
people that can offer necessary assistance in maintaining
quality in the distribution channel. As participation becomes
more synchronized, less money will be spent investigating the
causes of damage. . .the informationwwill already be available

in a database.

The system should include the participation of dealers and the
ultimate consumers where the installation occurs, because this
is often where concealed damage is finally discovered. In
addition to helping the manufacturing firm obtain damage
information, channel members can benefit from being a
participant in the quality system. By using guidelines and

standards to compare performance, they will be able to control

their processes as well.

 

There is not any one person that alone can control quality in
distribution. Driving the quality system should be a team,
including someone from Packaging, Traffic, Quality, and
Product Engineering. When needed, managers, foremen, carriers
and others can be brought in. All of the above have a common
goal: to reduce damage. Input from all is necessary when
solving logistical damage problems, because alone, each lacks

knowledge about the other areas.

65

There has to be someone to manage the team that manages the
system. Members of the above areas have specific
responsibilities that could limit them from being the overall
controller. Therefore, it is recommended that a mum
93W be appointed, someone who
oversees the damage management system and assures
functionality and progress. This manager should have contact
with all departments involved in distribution operations, as
well as channel members and carriers. The DOC Manager could
operate under similar formal procedures as the Production

Quality Control Manager does.

The DOC Manager should report to Logistics, and his/her
position would parallel that of someone responsible for other
logistical customer service activities such as order

management, warehousing, product support, etc.

The team, under the direction of the DOC Manager, should
define its mission and scope of interest. It should set
standards for procedures and product quality. If left to the
individual functions, conflicting performance goals will
result in standards that are difficult for some to comply
with. Standards should be simple to understand and follow,
then nonconformances will be easy to recognize. Diagrams and

graphics can help workers understand procedures, damage types,

66
and causes of damage without spending too much time reading

detailed reports.

After standards are set, the team can then implement methods
of monitoring conformance with them. Examples of methods were
discussed in Chapter Four. An information system is required
to formally feed back details of monitoring to those who can
analyze the data and correct causes of damage. In addition,
the same methods of monitoring processes and feeding back data
should be used to monitor any corrective actions, to see if

they improved the situation.

The communication of information is the vital link that ties
the process together. Because EDI can span the boundaries of
firms who own, handle and transport the product, it is the
quickest method of feedback, and can tie damage information to

other logistical information.

EDI can assist in collecting damage information because it
facilitates standardizing and formalizes monitoring. It is
already commonly being used for other logistical information
activities like customer service, order information, delivery
status, etc. The opportunity exists to include information

about the quality of products throughout the channel and at

67
delivery. As stated in RESULTS, several firms are already
looking into the possibilities of utilizing EDI to transmit,
and manage, damage information. Ethan Allen could also
benefit by adding EDI to their damage management system.
Currently, they transmit data by paper and telephone, and have
to spend time with data entry. With EDI, information could be

in the system and available for analysis immediately.

The opportunity also exists with EDI to include customer
information such as unloading capabilities and special
handling restrictions. As discussed, each of the firms voiced
a concern about the lack of knowledge in this area. Using
EDI, special packing and delivery, or unloading instructions

could be included with each order.

Through EDI, damage information can be transmitted to a
database. Part of the development process for the database
should include defining what kind of information is necessary
to make intelligent decisions. Features will vary between
firms, but regardless of the actual structure, the database
should be capable of manipulating data in many ways, relative
to who needs it. Some possibilities for analysis are:
frequency of occurrence of a particular product number, damage
as a percent of sales, and dealers experiencing the most
damage monthly. Analysis methods should help to determine

whether the cause of the problem is related to product

68
quality, handling procedures, transit environment or any other

inputs to damage.

One of Cavinato's premises, listed in Chapter One, was that
management often views loss and damage as a tolerable cost
that is not worth reducing or eliminating because the cost of
such effort appears greater than the benefits to be received.
However, after the initial implementation costs, EDI can save
money by eliminating many unnecessary steps. Paperwork and
data entry time will be reduced, as will large phone bills
resulting from transmitting damage information and trying to
solve problems over the phone. The labor time spent solving
problems will also decrease, because relevant data will be
available immediately. In addition, people might be more
willing to share information if it is easy to

do so.

£9§I£BENEEII.ANALI§I§

Unfortunately, this research does not include a cost analysis
for.implementing a distribution quality control system. It
was not possible to obtain information from the five firms on
how much damage costs them annually, and it was nearly as
impossible to determine the costs of implementation for each
of the firms, due to differences in size, channel structure,
etc. Consequently, comparisons could not be made. However,

potential benefits that could be realized follow.

69

 

Unfortunately, figures are not available to show the
difference between damage occurrences before and after
implementing a damage management system. There is no
numerical proof that having such a system can savetmoneyu One
can reason, however, that having reliable information about
damage and taking steps to reduce the sources of damage, will
lead to less dollars in lost profit due to damage, and may

increase sales.

Less money will be spent replacing merchandise, and paying
return transit costs. The benefits may vary by industry.
Higher charges are usually incurred when shipping replacement
product in a custom-built industry like furniture because the
lead time for completing the original order has been
shortened. The customer may need the product in a short
period of time (after they have already waited.the normal lead
time for the original product), and the sale may be lost
altogether. It is also possible that even more damage will
occur to merchandise being returned, thereby lessening its

potential to be salvaged.

IEPIQYQQ_EYELQE_§QREI21
A formal damage management information system will allow for

a continual understanding of occurrences in the distribution

70
environment that can contribute to damage. With improved
control over'the system, less time and money’will be necessary
to solve problems when they do arise, because valuable

information will already be at hand.

Q21$.EQ§!£&129_12_23259912§_H££2£1E1§

The Packaging Professional will be better able to optimize
packaging due to more complete information and better
knowledge of what problems and situations products are exposed
to in the distribution environment" This could result in less
packaging materials being purchased and used, based on the
knowledge that more is not always better. Overpackaging could

become a thing of the past.

t e a o d we Cus ome c os s
Damage loss, as well as costs associated with handling and
returning damaged merchandise are explicit customer costs.
Dealers and customers are a key part of a damage management
system and they should be involved in developing the system
that they will be held partly responsible for maintaining. By
exhibiting more concern to meet their needs, they might feel
more of a "connection" with the manufacturer, leading to

increased sales.

Customer service costs could decrease with the existence of a

damage information system. Less money and time for the

71
customer service representative (and those that further
analyze and try to correct problems) will be spent trying to
solve problems, because information will be readily available.

In depth searches for details of damage occurrences will no

longer be necessary in many situations.

 

Reliable information leading to better products and packages,
coupled with better knowledge for the workers who handle the
product, could lead to a significant reduction in damage, and
therefore claims. Better relationships with carriers could
most definitely exist with less controversy over who caused

damage.

W

Each of the case representatives agreed that the simple
knowledge that a formal damage management system exists could
be a benefit. Workers are more careful and helpful if they
are sensitive to damage issues, if they are educated to
observe and classify damage, and if they feel they are

essential to the company's success.

Workers' participation in the system should not be used
against them, where they are disciplined if they make an error
and cause damage. On the contrary, they should be encouraged

to pass on information about the causes of damage that they

72
experience so corrective action can be taken to avoid its
future occurrence. It is important to educate all employees
on the workings of the system, and how it can benefit the firm
as a whole. Not only should standards and procedures be made

clear, but also the reasons why they exist.

PAQEAQE_IE2LIQATIQN§

In addition to protecting the product, the package plays a
role in product modification, transport costs, and in the
claims settlement procedure. (A damage management information
system could also allow for the collection and analysis of
information that could help in these areas. However, the most
benefits realized will be in damage reduction due to better
information and therefore better packages. At last, the

packaging professional can know how much and what kind.

Solving quality problems which result from insufficient
packaging requires a method for predicting whether a proposed
package would perform better. Package protection performance
evaluation benefits a great deal from an information system

which improves knowledge about damage.

Using this information, tests can be correlated to real
performance through failure modes. This is known as the
"Damage Reproduction" theory of package testing. Developing

package test methods from damage information is a simple

73
matter of taking the available technology--state-of-the-art
package shock, vibration, compression, and shelf-life
evaluation equipment--and using it to reproduce damage. For
instance, if the corners of a particular piece of furniture
usually break off, there should be a test developed to judge

which package best prevents corner breakage.

The Damage Reproduction theory is used, for example, by the
United States Department of Agriculture who relies on the
Michigan State University School of Packaging to develop
performance specifications for the food it purchases. (It is
the largest food distribution system in the world including
Food-for-Peace, school lunches, prisons, charities, disaster
relief, and price support commodities). Performance
specifications are all test methods developed to reproduce the
damage most likely to occur in a given product-package-
distribution system, based on analysis of typical failures

(Twede et al, 1990).

Since there is no direct correlation between package cost and
performance, testing provides a means for comparing economical
alternatives to the package which is currently used. Thus,
even if there is currently no significant damage problem, a
damage information system can provide the direction for

testing to proceed by indicating prevalent damage modes.

74

IN_QQNQL!§IQN

Each firm investigated in this research would benefit by
combining their informal subsystems and developing a
comprehensive and formalized system to control product quality
in distribution. System development would begin by
establishing a team who can manage the system. That team
should first evaluate all those who will be a part of the
system to find out what part each will play. Next, everyone
involved should be educated and trained on the logic and
workings of the system. Procedures and product quality should
be standardized for simplicity, as should be methods of
transmitting data. The database containing damage information
should have the capability to analyze data in any ways
necessary to make intelligent decisions about corrective
actions. Corrective actions should be addressed at the causes
of the problems, not just the problems themselves. Lastly,
processes should be continually monitored to assure

conformance, even after corrective action has been taken.

A distribution damage management system can continually
identify lucrative opportunities to improve the quality of
products delivered. By setting performance standards,
tracking and analyzing conformance, and then upgrading
performance levels and re-evaluating standards, the logistical

quality system can continue to improve.

75

LIST OF REFERENCES

Anderson, June E. Case firm - personal interviews.

Baker, Edward M. 1988. "Managing Human Performance." innan's
' . Fourth edition. New York:
McGraw-Hill.

Banks, Jerry. 1989. ' c' s f ' o t . New‘York:
John Wiley & Sons.

Bennett, Don. Case firm - personal interviews.

Bingham, Gordon J. 1989. " '90' 3 Agenda Reveals Nod to Manage

Effectively in a Changing World." W 23 (Aug
14): 11- 12.

Boggiano, Chris. Case firm - personal interviews.

Bonoma, Thomas V. 1985. "Case Research in Marketing:
Opportunities, Problems, and a Process." W

Marketing_8esearsh 22 (May): 199-208-

Boulton, William R. 1985. "Case Study as a Research
Methodology-" Qa§e_8esearsh_lgurnal. 3-14-

Bowersox, Donald J., Closs, David J., and Helferich, Omar K.

1986. ,ngigt1§§1_ngngggmgnt. New York: Macmillan.

Bowersox, Donald J., Daugherty, Patricia J., Droge, Cornelia
L., Rogers, Dale 8., Wardlow, Daniel S. 1989. Lgnging
00- ee 8 ’ eule- v ‘es e! l. e_ _1‘ "O' .

11: Council of Logistics Management.

Burr, William G. Case firm - personal interviews.

Cavinato, Joseph L. 1975. "Analysis of Loss and Damage in
a Procurement Distribution System Using A Shrinkage
Approach." Doctoral dissertation, Pennsylvania State
University.

Crosby, Philip B. and Esterby, Jim. 1985. "Don't Forget
Distribution as Potential Problem Source." EHIQDQSIHQ-
(January 17): 66A31.

Crosby, Philip B. 1979. u ' 's e . New York: iMcGraw-
Hill.

76

Deming, W. Edwards. 1986. ,Qn;_gfnthg_§ni§1§, Cambridge, MA:
Massachusetts Institute of Technology.

Deming, W. Edwards. 1982. Qnglity, Egoductivity and
MW. Cambridge , MA: Massachusetts

Institute of Technology.

Feigenbaum, A. V. 1983. Igtgl Qnality angrol. Third
edition. New York: McGraw-Hill Book Company.

Fisher, Suzanne M. Case firm - personal interviews.

Geertz, Clifford A. 1973- W.

New York: Basic Books.

Gryna, Frank M. 1988. "Field Intelligence." gnngn;§_Qnglity
Control Handbook. Forth edition. New York: MoGraw-
Hill.

Juran, J. M. 1988. ' ' o d o . Fourth
edition. New York: McGraw-Hill.

Lambert, Douglas M. 1975. "The Development of an Inventory
Costing Methodology: A Study of the Costs Associated
with Holding Inventory." Doctoral Dissertation, The Ohio
State University. Chicago: National Council of Physical
Distribution Management.

Lansdale, David B. "The Vital Signs of Effective Packaging

Management. " Wasting-
"Leading Edge." 1990. Qgfl pistzinntion Ngtngnk Newsletteg.

4 (July): 1-2.

Lowery, Thomas. Ethan Allen examples provided through
personal conversations, 1991

Maezawa, Eiichi. 1987. "Product Modification to Reduce
Distribution Costs." SPHE_Janngl. (Spring): 21-22.
Meredith. LR 1987- WWW-

Third edition. New York: John Wiley & Sons. 512-551.

Miles, Matthew B. 1979. "Qualitative Data as an Attractive
Nuisance." ' ' t v S ' nc r e . 24
(December): 590-601.

Novak, Robert A. 1989. "Quality and Control in Logistics:
A Process Model." t o l

21W 19.
Porter, Michael. 1985. e ' v e: C d

A
\
\
‘.
\
\-
\

“van Maanen, John; Dabbs, J. M., Jr.; and.Faulkner, Robert R.,

77

Sueteining Superier Performance. New York: The Free

Press.
Radziewicz, Bruno. Case firm - personal interviews.

Robeson, James F., and House, Robert G., editors. 1985. The
Qietginntien Hangeook. New York: The Free Press.

Rogers, Everett M. 1983. Diffusion ef Innovatiens. Third
edition. New York: The Free Press.

Sinha, Madhav N., and Willborn, Walter O. 1985. 1113

Manegenent of Quality Aeeuranee. New York: John Wiley
& Sons. 472-497.

Thomas, L.F. 1965. The antrol ef Quelity. London: Thames
and Hudson.

Tribus, Myron. "Deming’s Redefinition of Management."
Qualiey Fireg. Paper, Massachusetts Institute of
Technology.

Twede, Diana. 1988. "The Process of Distribution Packaging
Innovation and Its Relationship to Distribution Channel
Structure." Doctoral dissertation, Michigan State
University.

l/Twede, Diana, and Goff, James W. 1989. "Distribution Damage

Measurement, Analysis and Correction."

Twede, Diana; Harte, Bruce; Goff, James W.; and Miteff, Steven
P. 1990. "Breaking Bags: .A Performance Specification

Philosophy Based on Damage." geurnel‘ 9f Peekaging
Teehnelegy. (Nov-Dec).

Twede, Diana, and Palmer, Caris J. "Logistical Loss and
Damage: Who Pays?" Institute Of Packaging Profeseionale
T h ’ rn l. Forthcoming.

eds. 1982. Vari§§i§§ of Qualitetive Reeeerch. Beverly
Hills, CA: Sage Publications.

vanSchoick, Charles. Case firms # personal interviews.
vasquez, Ray. Case firm - personal interviews.

venti, Ted. Case firm - personal interviews.

APPENDICES

78

APPENDIX A

W

setting standards:

Critical defects.are.defined, but threshold levels are not set
for judging acceptability limits. There are standards to
reflect common types of damage, but not causes, or costs
associated with damage.

Collecting and Analyzing Information:

1)

2)

Field Problem Reports. Used for damage communication,
shortages, overages, wrongly labeled items, and other
delivery and installation problems. Include time,
contact, order number, etc. Reason codes are used to
indicate particular problems. Reports also include
information on: "where," (ordering, plant,
distribution), "what," (product damaged, shortage/excess,
etc.), "why" (improperly loaded, inadequate packaging,
abusive handling, etc.), and types of "nonconformance"
(bent, chipped, dents, etc). .

FPRs are filed by field technical representatives who are
responsible to cover dealerships (broken down by region) .
They make sure the installation is going properly, and
give their opinions on damage, for example, whether it
needs to be replaced or can be field repaired.

Also filed by project managers, who are responsible for
the ultimate customer. Often, damage information comes
from them. They have to monitor progress on jobs, and
have a check—off sheet with carton numbers (which are on
every box), so they can see.noted.damage then, If damage
is noted, they call the customer satisfaction department,
who puts the information on an FPR.

FPRs can also be filed by a dealer.

FPRs cover external problems only (not problems on the
production line, for example). The representative
estimates that they get about 60% of damage information
this way.

Word of mouth. Field technical representatives call
packaging directly and tell them about problems. By
doing this they can expedite problems, and the
information is "straight from the horse's mouth." This
is helpful to the packaging department, because the
representative can send damaged product, pictures, and
information immediately. Dealers have to go through

79

customer service first, because they want to know up-to-
the-minute what's going on. If the dealers or customer
bypass customer service, the "left hand doesn't know what
the right hand is doing." (Field technical
representatives have the right to contact packaging
directly though.) Project managers gen call packaging,
but they must get approval through customer service
first. '

Letters from dealers are a source of damage information.

Dealer conventions offer a good opportunity to get first-
hand information.

An. estimated 35% of’ damage information is received
through the above word-of-mouth methods.

3) Survey carriers by phone each month to get opinions on
what they've seen.

4) Get some information from claims. If a carrier feels
something is a consistent problem, (and something they
-feel they aren't responsible for, especially concealed
damage with furniture), they'll raise the issue with the
company.

5) EDI is not fully'implementedn Damage information will be
included. It is tied to order information, but only
customer satisfaction has access to it, and quality uses
information obtained from them.

Corrective Action:

Corrective action reports (CAR). Controlled. by corporate
quality assurance, they are a tool for issue correction. 'They
identify issues, and a committee appoints the necessary people
and action for follow up (packaging, fix damage). Standard
operating procedures exist for taking corrective actions.
CARs can be used for processes and people: the problem doesn't
have to be damage related. Unfortunately, not many are filed
from field technical representatives since it is extra paper
work. This program is based on an earlier program that didn't
have "corrective" in the title or in practice. That program
wasn't effective because there was no corrective action and
nothing was being done. People were filling the forms out for
nothing.

Corrective action reports are more general than FPRs, covering
internal and external problems. Packaging sees them only if
they pertain to them (including damage). The company gets the
remainder of damage information through this program, which is
relatively new (not even a year old).

80

Discussion:
"Can't say that we hear about 100% of the damage, and not
until it becomes a real problem."

"Dealers have warehouses and can make a quick field
replacement, so the company does not hear about those problems
since they wouldn't.need.to»order'a replacement and.they don't
always return the damaged part".

"A lot of the damage is minor stuff (scratch, blemish) and
union installers can resolve a lot of the cosmetic blemishes
on the spot."

"Might miss some true reasons for damage, because the project
manager hires the installers, who also do the unloading, and
wants to maintain a good working relationship with them.
Therefore, the project.manager'might.not.pass on the fact.that
a worker actually dropped a box, or ran into it with the fork
truck, etc."

"Sometimes the project manager isn't completely educated on
specific problems with fragile products, or product that
should be handled in a certain way, so he/she cannot pass that
information on to the installers."

Not everyone has loading docks. The product could experience
a "good 4 1/2 foot drop off the back of the truck" if no one
is there to grab it. "Undoubtedly, there is a lot of damage
that takes place during the handling and unloading, and even
installing. Very often it has nothing to do with packaging,
but handling. Furniture is hard, it's bulky, heavy, and not
easy to handle quickly, especially when it gets more fancy.
Can't over-pack the.daylights out of it, because the.cost.gets
too high." -

"Packaging needs a foot in the door with product development
to give input on the structural integrity of things and how
they can survive the distribution environment. Input is
usually after-the-fact. Packaging gets brought in after
designs are frozen, and has a certain amount of the product
cost to work with for the package. It is necessary to develop
good relations with product development so they can begin to
understand packaging and handling needs, because it helps them
out in the long run."

Broad goals:

Corporate goal in 1990 to reduce FPR's by 50% overall.
Packaging did it.by 60%. Distribution as a whole reduced them
in excess of 50%. The same goal is in effect in 1991. Goals
put more emphasis on reports, where before they might have
been put on the back burner.

81

Benefits:
Damage has been reduced through value analysis in pkg. "There
is definitely a benefit to measuring damage."

"FPR's add a psychological benefit that makes everyone put
more care into their job. It's a management tool that you can
use to enforce what's got to’be done. It's follow up to track
progress. All people on the loading docks are fully aware of
how' many FPR's are processed against their department.
Management can track back to who actually loaded (the lowest
possible level), and get their involvement in a positive way
by asking how things could be different so the problem doesn't
happen again."

82

APPENDIX B

WM!

Betting standards:

Critical defects are not defined, and threshold levels are not
set for judging acceptability limits. There are no standards
to reflect controllable variables: common types and causes of
damage, as well as cost associated with damage.

Collecting and Analyzing Information:

1) Replacement orders (also called claims) are the major way

they get information. They covers "a good portion of
damage information." Packaging gets these replacement
orders from customer service when someone feels it is
something packaging could help with.
If packaging thinks there is a problem with something,
they tell customer service representatives to watch out
for replacement orders with that particular problem
.behind it.

2) Word of mouth information from phone calls, individual
dealers.

"Get more information than you'd think from carriers. The
customer makes a claim with the carrier. If there's a
problem, the carrier comes back to us, so you find out
‘damage information that way."

3) Postcards are attached to every order. The customer can
rate areas (on-time, etc.) including damage. The
customer order number is on the card, so problems can be
traced back to routes, carriers, etc. Postcards are
returned to shipping, who passes them on to packaging if
there are any poor ratings in damage, packaging, etc. , or
if there are any comments that could be helpful. Once
postcards are received (only about 10 at the most per
month), the packaging people just keep it in their minds
and look for recurring problems. The process is customer
driven - information is relayed only if they want to
respond.

Corrective action:
"We just do whatever has to be done. It doesn't matter if
it's not a packaging change."

Verbal follow up contact is usually made to find out what the
damage was really like.

83
Survey customers.

Training--Many workers don't understand principles of
unloading. They've found workers inverting loads from how
they were in the trailer. A big concern is that the customer
might not have equipment, loading dock, or trained workers.
"We need information on what the customer has available to
them...it should be included in order information."

Discussion:

Estimate damage at less than 2% of sales ("which is good").
Firm will soon be changing shipping terms to FOB destination
so firm B can handle claims instead of customers. This will
lead to better information.

The company doesn't see concealed damage claims sometimes,
because if installers can fix it or get it from a dealer's
warehouse, there is no replacement order.

Negative aspect of surveys and postcards: There might be one
damaged product out of thousands, and the customer will circle
"poor packaging". So you don't have quality information all
the time. "You don't know who's telling you what, and you
have no idea of the magnitude of the problem. There is a lot
of subjectivity involved."

Packaging could go to shipping and get a printout of every
cracked drawer front from here to California in the last six
months, but it would take forever because they can sort by
destination but then they have to read each comment section.
Plus, it often only says that damage existed in the shipment,
but not what products or specific type of damage occurred, or
how it happened.

Benefits:
There has been a reduction in damage since they started
monitoring just because packaging took action.

There is a psychological advantage from handlers knowing their
work is being monitored.

84

APPENDIX C

W

Setting standards:

Critical defects are not.defined, and threshold levels are not
set for judging acceptability limits. There are no standards
to reflect controllable'variables: loommon‘types and causes of
damage, as well as cost associated with damage.

Collecting and Analyzing Information:

1)

2)

3)

3)

Replacement orders.

Tracking' the :number' of complaints (or requests for
replacement parts), can give a good idea of what issues
continually come up.

Surveys from dealers.
Field audits at major installations.

Any order that is a certain amount or more gets audited
by questionnaire, and over another amount has a field
audit. Audit surveys are collected by the quality
department, and are used in quarterly reviews. An
estimated 20% of all orders are this amount or more, but
that 20% is 80% of the product volume. It was estimated
that they get about a 45% response rate from. the
questionnaires. 80% of orders are 20K or less,
consisting usually of replacement or addition product.
But, the big orders usually go TL, while the others go
LTL where most of the damage happens. So, audits usually
come back with minimal damage information.

Business partner in Japan distributes some product. They
receive and warehouse the product, and handle the

. Japanese logistics from there. Their standards are very

high, and they "won't take anything that's blemished in
any way." For this reason, product is inspected before
it's containerized. Very detailed weekly reports are the
result. Statistically significant information is
received, and it is considered to be very indicative of
what's going on with damage throughout the entire
distribution channel. "It gives you a good sense of
whether the package is doing its job or not." However,
by the time it's inspected before being containerized,
it's already been shipped 3000 miles. The reports often
claim that "damage probably occurred before packaging,"
so it could be from some point in the manufacturing
process. Their standards can be over-demanding.

85

4) "The company still depends a fair amount on informal
communication", and it's the best kind of information
sometimes, but you don't get it until it's already a big
problem."

5) Get. monthly' total damage reports (in. dollars) from
carriers, but no specific detailed information.

6) EDI. Real time on-line system via satellite. Major
dealers get a dish. They can check on the status of
orders, place orders, change orders, etc. The system is
evolving so dealers can include information about damage
and replacement orders.

7) Verbal communication from people that actually operate
the dealership (versus the owners). "Get lots of first-
hand problems and information that way. We need to
continue to get out there on a regular basis because
that's where you get a lot of information."

Corrective Action:

Quarterly quality reviews (from data gathered), highlight any
particular trends from a standpoint of quality. Then, the
product maintenance manager and packaging manager go through
and review the data with the quality department using trends
on various product types. They focus on the five biggest
quality problems each quarter and that's where they put the
major resources. Then they work to get it off the list, but
continue monitoring it.

Packaging teams are sent out to investigate specific problems.

The Packaging Manager is currently putting together a training
video for a packaging process. "Over time, workers get away
from the way it's supposed to be done". So, they're making a
video to remind.workers;gny;the standards were set that way in
the first place. The problem stems from conflicting goals,
because the workers have to keep pace with production
schedules. Packaging then has to go back through management
to suggest a change in the work standards. It is difficult
though, because management doesn't want to have to put an
extra worker in to keep up with the production pace.

Goal to reduce corrugated use, primarily by increasing blanket
wrap.

Discussion:
The Packaging department is working very closely with
distribution to increase use of TL shipments.

"Feedback of course is important"

86

Classes of dealers: Some are committed to their product, and
other, non-competing lines are approved. They are called
office pavilions, and.there are about 35 of them. 'The company
does "a good deal of our business through them." The company
handles their payroll, logistics planning, inventory, and
other overhead issues.

The next level of dealers are a little more independent. They
won't agree to all of the stipulations required to be a top
level dealer. They could do as much of half of their volume
through the company. They maintain a little more autonomy.

The person placing the reorder often isn't the person who saw
the damage. They were just told to order a new one.
Information is second- or third-hand. The hey of the damage
is hard to get. "Anytime you can get communication closer to
your agents, the opportunity to get the he! improves"

There is a lost opportunity when inventory is taken monthly
and quarterly at a warehouse because damaged product is
scrapped out, and it doesn't say why (there are no detailed
reports).

Quality ‘teams (deal with. specific jproblems (for’ example,
products, on-time (delivery, etc.). The team. is 'multi-
disciplined. "For example, there's a team working on
distribution techniques, trying to get away from using LTL.
I may be being unduly harsh, but I think furniture and LTL1are
incompatible, and I'm pushing the organization to get away
from it.

87

APPENDIX D

W

Betting standards:

Critical defects are not.defined, and.threshold.levels are not
set for judging acceptability limits. There are no standards
to reflect controllable variables: common types and causes of
damage, as well as cost associated with damage.

Collecting and Analyzing Information:

1)

2)

3)

4)

5)

'National accounts - the company files claims for these
accounts, and can gather damage information from them.

Special claim investigations.

Carrier monthly reports - They show the claim amount, the
type of claim, the product (which is not always listed),
the status of claim, and whether’it.is concealed or noted
damage, but not what kind of damage. Carrier monthly
reports are analyzed more often to find out what carriers
are getting more damaged product (so as to correct
problems with the carrier) than they are analyzed to find
out what products experience more damage (unless it's
noted that a particular product is experiencing a lot).
Packaging is a minor problem. The company goes beyond
classifications. They rarely have carriers denying
claims due to improper packaging.

Mostly word-of—mouth the ' e 't'v rob

Then people in other departments are called in, and they
do quality assurance and packaging checks. "It all
centers around the information that we get back from our
dealers." There is also informal communication from
people from various departments when there is a problem.

Expediter's monthly report. Covers problems that are
noticed before product gets out the door. These reports
are more detailed than information about finished goods
in the channel. Paperwork is generated to send product
back for rework in the area where there was a problem.
It helps to assure the product will not go out the door
damaged. Damage has gone down since they've been
measuring, maybe because workers know the report is being
generated. "Before it's packaged, it should not be
damaged. If it is, they should be sending it back.
Unfortunately, some things do get passed on because they
figure someone on down the line will take care of it."

88

6) Surveys. Questionnaires are sent out on all aspects of
distribution (on-time, shipment complete, friendliness of
drivers, etc.) The firm doesn't get a lot of feedback,
and they aren't detailed about damage.

7) Audit teams sent out in the field will find out a lot of
information, but this isn't done often.

Corrective action:

Educating dealers on how to unload is very important in being
preventive, because it's really a problem, especially since a
lot of the workers at smaller“warehouses are summer'help, part
time, etc. like a grocery stock boy. "I predict that most of
these claims are an issue of unloading. It's a gut feeling,
but we often find problems there." But it's a lot of money to
make videos, etc.

Discussion:

"We don't really know damage in dollars. Could tally (very
time consuming) national account claims from monthly reports,
and get outstanding claims, but there is nothing in relation
to percent of sales, etc. National accounts are a very small
portion of total business though."

"As a corporation we're very serious about damages." "We're
sure that we're missing some of the claims on that data,
because of who's filing the claim."

"We have less damage because the majority of what we haul is
blanket-wrapped." This ties in with using considerably more
TL shipments.

The firm has the capability on computer to sort by product and
find out how'many times it appeared in claims (but only if the
claim specified the product). All information is mixed in
with the company mainframe, and it's "certainly not the total
picture, because it's just the claims we get involved in."
These claims are called dealer assistant claims, and arise
when the carrier doesn't pay. The firm will work with these
dealers, and investigate the problem to get something out of
carrier.

Year-end reports have the total of claims filed, but the
number includes loss, damage, delay, theft. They are not
separated.

The worst claims are for file cabinets (because of weight,
awkward size). If there is a particular series or line of
cabinets getting damage, they go back.to the source and try to
figure out where/why it happens.

89

"When you look at the product as a whole that we manufacture
and ship out of here, in number of pieces, we don't really
have a serious damage problem. If you compare the number of
pieces damaged with the number of pieces we ship out of here
in a year, its really not that bad. But when damage does
happen, we've got a customer'waiting for it, and now'they have
to wait some more, and sometimes it gets damaged a second
time."

"Corporations are reluctant to publicize a negative issue.

There is less damage with product shipped uncrated (blanket-
wrapped), which is 63% of TLs, or volume shipments. However,
this only represents about 40% of the product. Concealed
damage goes awayu Can inspect and reject products right away.
This helps with product coming off the line too, since it can
be rejected before shipment.

"There are a lot of products that we are forced to package
(glass, lights, wood table tops, etc.). We would like to ship
everything we can uncrated. It depends on the destination,
because some dealers aren't set up to store uncrated product.
More product can go uncrated if it's going direct to the job
site."

"Preventive maintenance is tough to deal with in a corporation
because upper management doesn't want to extend any money.
You have to justify it, but where's your proof? There's no
justification in numbers."

Benefits: Less delays and callbacks.
Psychologically, workers might handle product better.

90

APPENDIX E

W

Betting Standards:

Critical defects are not defined, and threshold levels are not
set for judging acceptability limits. There are no standards
to reflect controllable variables: common types and causes of
damage, as well as cost associated with damage. The system is
currently under revision to include such variables.

Collecting and Analyzing Information:
Customer service is the recipient of the data, and quality
assurance collects it and manages the database. They then
publish quarterly reports with trends.

1) Customer service replacement orders. Normally, if there
is packaging related damage, the dealers (who are
installing) will get back to customer service. Ideally
the company will get a replacement order every time there
is damage because the customer needs new product.
Customer service notes why there is a replacement order,
so you know about damage. "They give a good indication
of the overall damage problem." It was hard to implement
the system, because sales people, customers, etc. don't
want to do any more paperwork than absolutely necessary.

2) Pre- and post-pack audits.

3) Internal document called Discrepant Material Report (DMR)
can also give damage information from stock situations.
Would not get much damage-due-to-packaging information,
mostly handling problems. Even so, the document
indicates there was concealed damage, but not what kind
or how it happened.

4) Electronic mail - all major dealers, sales offices, and
showrooms are tied in. Good method of obtaining damage
information because it's instantaneous.

Discussion: .
There "really hasn't been a big damage problem", estimated at
less than 1% of sales.

"If it's somebody else's fault (like someone driving a fork
truck.through a file), we're not going to hear about it." But
if it's noticed right when they break the seal on the truck,
or remove the blanket wrap, they'll file an FPR right away.

91

"If there's stock at the branch, they can wheel and deal
through the branch/warehouse, and you never find out about
some damage." However, now the company requests any
questionable product back so they know what's happening.

Corrective actions:
Quality teams focus in on specific problems.

Track non-conformance costs through FPR system. Problems get
traced back using problem codes to a work center or
department. It then affects their budget as a cost allocated
against that work center. They are evaluated on a monthly
basis.

They used to have a quality document that was sent with each
product (quality service card). It was discontinued because
they were going to update the way they collected data because
the information was too subjective.

"Quality is in the eyes of the consumer."

"Tillilfljllllfllflflfllfllflr