MSU ’ LIBRARIES ._:_. RETURNING MATERIALS: P1ace in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. 't4<{.? iogs pg AN EMPIRICAL FIELD STUDY OF THE ROLE OF COST ACCOUNTING IN A COMPUTER-INTEGRATED MANUFACTURING ENVIRONMENT by Darrel Irvin Gosse A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Accounting Graduate School of Business Administration 1989 ABSTRACT AN EMPIRICAL FIELD STUDY OF THE ROLE OF COST ACCOUNTING IN A COMPUTER-INTEGRATED MANUFACTURING ENVIRONMENT By Darrel Irvin Gosse A field study was conducted to contrast the role of cost accounting different in four computer-integrated manufacturing (CIM) plants with four traditional (batch-oriented) manufacturing plants. Empirical data was gathered from on-site interviews with accountants, engineers and production personnel. Empirical findings were derived from interview transcripts and plant tour notes. Four research hypotheses were investigated, one for each of four activities the comprise the defined role of cost accounting: cost iden- tification, cost entry, cost assignment, and cost reporting. It was hypothesized that CIM plants would designate more (but smaller) cost centers, would group their cost accounts by class of resources (mater- ials, workers, machinery and equipment, tooling, technology and informa- tion, and facilities), and would establish separate overhead pools for each class of resource. The CIM sites were expected to choose different resource events as cost transaction triggers for cost entry. Cost assignment at CIM sites was expected to lead to sub-dividing of overhead pools and use of more specific allocation measures tailored for each class of resource. Cost reporting at CIM sites was expected to conform to a weekly reporting cycle, like that production, and generally focus on strategic manufac- turing objectives (delivery reliability, production flexibility and consistent high quality). Fourteen related ”hypothesis focus points" were developed from background literature to aid the process of analyz- ing the empirical findings. Three of the four hypotheses were weakly supported by study find- ings and one (cost entry) was not supported. Less contrast between CIM sites and traditional sites was observed then expected. However, the information obtained through analysis of the hypothesis focus points provided many useful insights into the role of cost accounting. Copyright (C) by DARREL IRVIN GOSSE 1989 DEDICATION snwaan 0033: (1911 - 1952) Hy father, who has inspired me far more than he could have imagined. iv ACKNOWLEDGMENTS I express my deep appreciation to the four professors who served on my dissertation committee: Professors Harold M. Sollenberger (Chairperson), David M. Dilts, Susan F. Haka, and Steven A. Melnyk. These four stalwart individuals encouraged, nudged, and helped direct this project to a successful conclusion. Much appreciation is also extended to the National Association of Accountants for a generous research grant which helped to fund the project. AN EMPIRICAL FIELD STUDY OF THE ROLE OF COST ACCOUNTING IN A COMPUTER-INTEGRATED MANUFACTURING ENVIRONMENT TABLE OF CONTENTS W W I -- INTRODUCTION II -- RESEARCH PURPOSE . THE ROLE OF COST ACCOUNTING Cost Identification . Cost Entry . Cost Assignment . Cost Reporting COMPUTER-INTEGRATED MANUFACTURING -- "CIM" CONTRIBUTION OF THE STUDY LIMITS OF THE STUDY CONTENT OF OTHER CHAPTERS RESEARCH CONTEXT . MANUFACTURING: A MATTER OF RESOURCE MANAGEMENT . Responsive Flexibility Quality by Design . Delivery Reliability Cost Management . . . The Cost Management Challenge . CONTRASTING TRADITIONAL MANUFACTURING AND CIM IMPLICATIONS FOR COST ACCOUNTING ACTIVITIES Revised Categorization for Identification of Costs Changes in Data Collection . Cost Assignment Based on Resources Cost Reporting and Cost Objectives Cost Accounting Activities Summarized . Defined Terms . SUMMARY vi xii xiii w ooosnbw 14 15 15 17 17 l7 19 20 21 22 23 25 26 32 34 38 40 41 42 III -- BACKGROUND FROM RELEVANT LITERATURE . INTRODUCTION LITERATURE ON TRADITIONAL COST ACCOUNTING . TRADITIONAL MANUFACTURING DEFINED HISTORIC ORIGINS OF TRADITIONAL COST ACCOUNTING Emergence of Labor-Oriented Costing . Early Standard Cost Systems . . Emergence of Factory Burden . Creation of Cost Centers Attaching Costs to Products . STANDARD COST SYSTEMS . . . Harrison (1921 and 1930) Henrici (1960) DeWelt (1975) Standard Costs and Variance Analysis (N. A. A. 1974) THE ROLE OF THE CONTROLLER . . . Accounting Activities for Reporting Cost Data . Impediments to Acceptance of Cost Data Cost Data Measured in Physical Terms SUMMARY: TRADITIONAL COST ACCOUNTING LITERATURE LITERATURE ON CIM AND ITS IMPACT ON COST ACCOUNTING . LITERATURE ON CIM CHARACTERISTICS CIM: Characteristics and Advantages . CIM: Influence on Cost Accounting . RECENT COST ACCOUNTING LITERATURE CAM-I (1985 and 1986) . Cooper and Kaplan (1988) Dilts and Russell (1985) Johnson and Kaplan: "Relevance Lost" (1987) Johnson (1987) . . . . . . . . . . Kaplan (1983,1984a, 1984b) Kaplan (1985a) Miller and Vollmann (1985) . . Summary. Recent Cost Accounting Literature vii 43 43 44 44 46 47 47 48 48 49 51 51 53 53 54 55 56 57 59 59 6O 61 62 62 64 65 7O 71 72 74 75 76 78 79 RESEARCH HYPOTHESES AND FOCUS POINTS . . . . . . . . . . . . . . 80 COST IDENTIFICATION . . . . . 80 Redesignation of cost centers (81); Classification of resources and support (81); Identification of cost manage- ment responsibility (82); Classification of direct costs (83) COST ENTRY . . . . . 83 Placing triggers close to resource events (84):, Respon- sibility for cost data -- a sense of data ownership (84); Triggers for expendable and durable resources (85); Keep- ing cost data current (85) COST ASSIGNMENT . . . . 86 Assigning support costs with relevant resource factors (87); Assigning costs based on cost management objectives (87); Penetrable costs with transparent cost assignment (88) COST REPORTING . . . . . 88 Cost objectives and strategic manufacturing objectives (89); Cost management reporting for cost monitoring (89); Reporting cost consequences of disruption and waste (90) SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 IV -- RESEARCH DESIGN . . . . . . . . . . . . . . . . . . . . . . . 92 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . 92 RESEARCH DESIGN OBJECTIVES . . . . . . . . . . . . . . . . . . . 92 THE DESIGN CHOICE -- A FIELD STUDY . . . . . . . .h. . . . . . 94 DESIGN GUIDELINES FOR THE FIELD STUDY . . . . . . . . . . . . 95 Site Selection . . . . . . . . . . . . . . . . . . . . . . 95 Classifying the Sites . . . . . . . . . . . . . . . . 97 Giving Focus to the Field Inquiry . . . . . . . . . . . 98 Objectives for Data Gathering and Analysis . . . . . . . . 99 FIELD RESEARCH PROCEDURES . . . . . . . . . . . . . . . . . . . . 100 CRITERIA FOR SITE SELECTION . . . . . . . . . . 100 Selection of the Firms -- Four Comparative Pairs . . . . . 101 THE SITES DESCRIBED . . . . . . . . . . . . . . . . . . . . . 102 Abnett -- (CIM Site) . . . . . . . . . . . . . . . . . . 103 Bladnu -- (Traditional Site) . . . . . . . . . . . . . . . 105 Ceston -- (CIM Site) . . . . . . . . . . . . . . . . . 105 Dolnar -- (Traditional Site) . . . . . . . . . . . . . . . 107 viii Elnep -- (CIM Site) Flaxtin -~ (Traditional Site) Gledbul -- (CIM Site) . Holpin -- (Traditional Site) . Summary of the Selected Sites . FIELD INTERVIEW GUIDE Decisions Related to the Cost Identification Hypothesis . Determining the work center schedule (117); Choosing routings for production (117) Decisions Related to the Cost Entry Hypothesis . Undertaking a cost reduction analysis (119); Deciding to change manufacturing processes (119); Changing methods for setting or revising standard costs (120) Decisions Related to the Cost Assignment Hypothesis . Establishing overhead application rates (121); Developing quotes or bids for new business (121) Decisions Related to the Cost Reporting Hypothesis Revising end-of period transaction cut-off practices (122); Revising the form or frequency of cost accounting reports (123) SITE VISIT PROCEDURES Advance Arrangements Interview Procedures at Site Visits . POST- VISIT DATA ANALYSIS PROCEDURES . Transcription of Interview Tapes Search Through Transcriptions for Findings Analysis of the Specific Findings . SUMMARY . V -- THE EMPIRICAL FINDINGS INTRODUCTION SALIENT FINDINGS Cost Accounting as a Function Should Provide More Leadership . Manufacturing' s Technological Resource Events are not Well Specified . Direct Labor is not an Effective Surrogate for Technological Resources . ix . 108 . 110 . 110 . 111 . 112 . 115 . 117 . 118 . 120 . 122 . 124 . 124 . 125 . 126 . 126 . 127 . 127 . 128 . 129 . 129 . 130 131 . 132 . 133 SUPPORT FOR THE RESEARCH HYPOTHESES . Hypothesis Hypothesis Hypothesis Hypothesis ”WNH ANALYSIS OF HYPOTHESES FOCUS POINTS . COST IDENTIFICATION HYPOTHESIS . . 1.1 REDESIGNATION OF COST CENTERS . . . 1. 2 CLASSIFICATION OF RESOURCES AND SUPPORT . 1. 3 IDENTIFICATION OF COST MANAGEMENT RESPONSIBILITY 1.4 CLASSIFICATION OF DIRECT COSTS Summary -- Hypothesis 1 Focus Points COST ENTRY HYPOTHESIS . . . . . 2.1 PLACING TRIGGERS CLOSE TO RESOURCE EVENTS . . 2. 2 RESPONSIBILITY FOR COST DATA -- A SENSE OF DATA OWNER- SHIP . . . . 2. 3 TRIGGERS FOR EXPENDABLE AND DURABLE RESOURCES . 2.4 KEEPING COST DATA CURRENT . . . . Summary -- Hypothesis 2 Focus Points COST ASSIGNMENT HYPOTHESIS . . 3.1 ASSIGNING SUPPORT COSTS WITH RELEVANT RESOURCE FACTORS 3.2 ASSIGNING COSTS BASED ON COST MANAGEMENT OBJECTIVES . 3.3 PENETRABLE COSTS WITH TRANSPARENT COST ASSIGNMENT . Summary -- Hypothesis 3 Focus Points COST REPORTING HYPOTHESIS 4.1 COST OBJECTIVES AND STRATEGIC MANUFACTURING OBJECTIVES 4.2 COST MANAGEMENT REPORTING FOR COST MONITORING . . 4.3 REPORTING COST CONSEQUENCES OF DISRUPTION AND WASTE . Summary -- Hypothesis 4 Focus Points . . . . . . SUMMARIZING THE FINDINGS FOR CIM AND TRADITIONAL FIRMS . Cost Identification -- CIM vs. Traditional Cost Entry -- CIM vs. Traditional . Cost Assignment -- CIM vs. Traditional Cost Reporting -- CIM vs. Traditional . SUMMARY OF SUPPORT PROVIDED BY THE EMPIRICAL DATA CONCLUSIONS . VI -- CONCLUSIONS AND OPPORTUNITIES FOR FURTHER RESEARCH . INTRODUCTION . 135 . 136 . 138 . 140 . 142 . 144 . 145 . 145 . 151 . 154 . 157 . 159 . 160 . 161 . 165 . 168 . 169 . 170 . 171 171 . 175 . 178 . 182 . 183 183 . 187 . 192 . 194 . 195 . 196 . 197 . 199 . 200 . 201 . 208 . 209 . 209 DISCUSSION OF THE HYPOTHESES IN RETROSPECT COST IDENTIFICATION General Conclusions . Specific Suggestions Broadening the focus to a wider set of resources (211); Developing improved specification of technological resources (212); Designating cost centers for product- focused cells (212) COST ENTRY . . . General Conclusions . Specific Suggestions Replenishment events as transaction triggers (215)', Trig- gering events of usage of durable resources (215); Trig- gers that capture process ”pay points” (215) COST ASSIGNMENT . . General Conclusions . Specific Suggestions . Classifying overhead pools to distinguish the nature of the costs (217); When assigning costs, match support costs with resources receiving the support (217); Measure the usage of capacity resources to assign capacity costs (218) COST REPORTING . . General Conclusions . Specific Suggestions . Cost reporting to support cost management objectives (221); Encouraging resource managers to use cost system as a primary source of cost data (222) SUGGESTIONS FOR FURTHER RESEARCH ADDITIONAL FIELD RESEARCH ANALYTICAL MODEL RESEARCH FIELD EXPERIMENTAL RESEARCH CONCLUSION W I -- FIELD INTERVIEW GUIDE II -- FIELD VISIT ARRANGEMENT LETTER . WES - INDEX xi . 209 . 210 . 210 . 211 . 213 . 213 214 . 216 . 216 . 217 . 218 . 218 . 221 . 223 . 223 . 224 . 224 . 225 . 226 . 249 . 250 . 256 10 11 12 LIST OF TABLES Comparison of Manufacturing Objectives . Comparison of Resource Account Groupings . Comparison of Cost Objectives Comparison of Cost Accounting Activities . Comparisons of Individual Pairs of Firms . Pairings of Sites Visited Summary of Characteristics of Selected Sites . Manufacturing Decisions and Hypotheses . List of Hypotheses Focus Points Composition of Overhead Pool at One Traditional Site . Summary of Support for Hypotheses Summary of Support for Hypotheses Focus Points . xii 24 29 39 4O . 101 . 103 . 113 . 116 . 145 . 172 . 202 . 204 LIST OF FIGURES Comparison of Floor Lay-outs, CIM and Traditional Plants . Comparison of Transaction Trigger Points . Patterns of Cost Assignment Alternative Patterns of Cost Assignment xiii 3O 33 35 37 CHAPTER I INTRODUCTION Prior to the mid-19708, U.S. manufacturers competed successfully against manufacturers from other industrialized nations. Then global competition intensified. U.S. manufacturers began to lose market share of large volume products to foreign imports, including motor vehicles and components, electrical and electronic products, consumer appliances, wearing apparel, machine tools, and metal forming equipment [Scott and Lodge (1985)].1 Initially, U.S. manufacturers responded to foreign competition by focusing their factories [Skinner (1974)], aligning their manufacturing processes with product characteristics [Miller (1981)], or by moving their production off-shore to nations with lower labor rates [Reich (1983, ch. VII)]. But eventually these strategies failed to stem the tide of foreign competition. Then, U.S. manufacturers began to examine foreign competitors' strategies and practices, which led to changes in the organization and technology of manufacturing processes [Limprecht and Hayes (1982); Wheelwright (1981)]. Responding to foreign competition, U.S. manufac- turers began to invest heavily in computer-aided technology used for en- gineering design, production scheduling and computer-aided machinery. Manufacturers used computer-aided technology to integrate manufacturing processes to a greater extent than typically possible in traditional manufacturing [Gunn (1982)]. This combination of computer-aided tech- nology and integrated manufacturing processes, or "computer-integrated 1For a specific analysis of U.S. manufacturers decline in market share of manufactured products during the 1970s, see Ketterling (1984); other sources include Abegglen (1983), Fallows (1980), Hayes and Aber- nathy (1980), Hayes and Wheelwright (1984). manufacturing" (CIM), has changed the context in which cost accounting operates. This study will look at the contrast between two types of manufac- turing: traditional and CIM. ”Traditional manufacturing” is defined as batch-oriented production that focuses on economies of scale and is organized around process traits [Reeves and Turner (1972)]. "CIM“ is defined as computer-aided production, with integrated processes, designed and operated to achieve economies of "scope” with the cost efficiencies of economies of scale [Goldhar and Jelinek (1985, 1986)]. Both traditional manufacturing and CIM are defined in greater detail in Chapter II. CIM is organized and managed differently than traditional manufac- turing because the manufacturing strategies and process characteristics are different. In traditional manufacturing, products are generally standardized, efficient production lot sizes are large, product life cycles are long, and production technology is stable [Buffa (1984)]. In CIM, manufacturing processes are more integrated and more flexible, products have more variety, efficient lot sizes are smaller, product life cycles are shorter, and technology changes are more frequent [Gold- har and Jelinek (1983); Gunn (1982)]. CIM enables manufacturers to respond quickly to changing market demands and yet attain high levels of quality and delivery reliability [Buffa (1984)]. CIM also changes the nature and patterns of cost incurrence, which has an important bearing on the role of cost accounting. Most contem- porary cost accounting concepts and practices are based on an implicit model of traditional batch-oriented manufacturing that may not correspond with the characteristics of the CIM environment [Kaplan (1984b)]. Recent literature suggests that traditional cost accounting systems may no longer be adequate for managing costs. For example, a recent Fortune article says "most companies are working with flawed estimates of what it costs to make their products" [Worthy (1987, p. 1.3)].2 RESEARCH PURPOSE This research explores four activities that comprise the role of cost accounting: cost identification, cost entry, cost assignment and cost reporting. The purpose is to investigate changes in these activi- ties when CIM is introduced. The research question is: How does the integration of manufacturing processes and the application of computer-aided technology affect the role of cost accounting in cost management systems? This question is investigated with four research hypotheses, one for each cost accounting activity. The role of cost accounting and the related hypotheses are specified in the next section. CIM is defined for purposes of the study in a later section of the chapter. The research context is described in greater detail in Chapter II. THE ROLE OF COST ACCOUNTING Cost accounting identifies, measures, assigns, and reports costs to managers so the managers can be cost effective in their use of resources 2For examples of recent literature concerning cost accounting systems in CIM environments, see Johnson and Kaplan (1987), Johnson (1987), Kaplan (1983, 1984a, 1984b and 1985a), Dilts and Russell (1985), Johannson (1984 and 1985), Miller and Vollmann (1985), N.A.A. (1986), Seed (1984) and Wingard (1985). to achieve manufacturing objectives. The basis of cost measurement is a "resource event," which is an instance where an "expendable” resource is consumed or a "durable” resource is used. A "cost” is the result of the consumption or use of resources for the production of other resources, or for support of such production. "Waste" involves the consumption or use of resources but, unlike "cost,“ does not produce other resources nor support production and does not maintain, support or enhance product value. "Cost management" is a strategic manufacturing objective. To manage costs is to use cost data in making decisions that involve the consumption or use of resources and thereby control the costs that result. Cost accounting's ”role" is to provide cost data for purposes of cost management; this role is comprised of four activities: cost identification, cost entry, cost assignment and cost reporting. MW Cost accounting begins by identifying costs as resources consumed or used. Since individual managers are accountable for resources pro- vided to them and for the costs that result from resource events, cost identification also includes determination of manager accountability. "Cost identification" is the process of recognizing and classifying costs based on the nature of resources consumed or used and on the specific management responsibility for the resource consuming or using event. Cost identification includes: (1) establishing cost accounts, (2) designating cost centers, and (3) determining levels of management responsibility for costs. Cost recognition involves certain fundamental definitions. "Resources“ are materials, workers, machinery and equipment, tooling, technology and information, and facilities consumed or used in produc- tion or to support production. Cost accounts may be grouped by class of resource to reflect the nature of resource events that result in costs. To "consume” a resource is to use it entirely as a discrete event (e.g. to use an expendable fastener or apply lubricants). To "use“ a resource is to consume it in a gradual series of events (e.g. repeated applica- tions of a fixture, mold, or die). "Direct" costs follow from outright conversion of input resources to output resources; "indirect” costs involve supportive resource events that precede or follow direct conver- sion. Costs are recognized where resource events are measurable and assignable to responsible resource managers, and costs are classified so they can eventually be assigned and reported according to cost objec- tives. Effective cost identification requires a logical, natural clas- sification scheme with appropriate names for accounts, logical account groupings, correct definition of cost centers, and other means of bound- ing cost responsibility. All of these are determined by "cost objec- tives” -- purposes for which costs are reported, or end uses of cost data. The term ”boundaries of cost responsibility" broadly describes cost responsibility recognition, cost account specification, and cost clas- sification. Cost centers are established for collections of resources managed by individual managers. Classification includes a determination of levels of cost responsibility. High level managers are responsible for some costs, such as production planning and control, while lower level managers are responsible for other costs, such as repairs or rework. CIM changes traditional patterns of resource events, which implies changes in (1) how cost responsibility is recognized, (2) the names used for cost accounts, and (3) how accounts are grouped for cost assignment and reporting. Cost centers may be expected to be redesignated and accounts regrouped. CIM production processes are organized according to product traits, while traditional manufacturing organizes resources according to process traits [Goldhar and Jelinek (1983)]. Large spe- cialized process departments of traditional manufacturing are likely to be replaced in CIM with smaller, product-focused process areas, called cells [Groover and Zimmers (1984)]. Therefore, cost centers in CIM plants are likely to be smaller in size (floor space occupied) but more numerous than in traditional manufacturing. Cost centers will be smal- ler in size because product-focused resource clusters occupy less square footage of floor space than traditional larger specialized process areas. Smaller cost centers are needed so that the costs and activities within the cost centers are reasonably homogeneous. Traditional cost accounting systems classify production costs as either materials, labor, or overhead. Manufacturing processes in CIM are more dependent on computer-aided technology, more integrated, and generate significant resource support costs. Traditional cost classifi- cations as materials, labor, and overhead do not adequately define the resource set used in CIM. Accordingly, resource classifications in CIM may be divided into more groupings, such as materials, workers, machinery and equipment, tooling, technology and information, and facil- ities. As discussed subsequently under cost assignment, CIM provides the opportunity to match support costs with related resources and assign combined resource and resource support costs to cost objectives. Cost identification in CIM requires, therefore, that accounts be grouped by class of resource. Based on theory developed in later chapters, the ”Cost Identifica- tion“ hypothesis is: Hypothesis 1: Compared with the traditional setting, CIM will lead to a greater number of cost centers, which will be smaller in size. In addition, classification and groupings of ac— counts will be based on a wider set of resource groups than materials, labor and overhead. The “Cost Identification" study area investigates this hypothesis. QQ§&_EDEII Certain resource events are selected as cost transactions for collection and entry of costs into accounting data files. Resource events are selected to provide objective, reliable measurements and satisfy cost objectives. The resource events that are selected for cost transactions are ”cost transaction triggers.” CIM's resource events are likely to be different than those of traditional manufacturing. CIM has a variety of potential trigger events, including entry and exit events, time-oriented events, and events that affirm conformance to specifications. While many of these events also occur in traditional manufacturing, they can be measured more precisely in the data-rich CIM environment. The CIM data collection systems can be expected to take advantage of these and choose different transaction triggers, thereby making cost accounting systems a more integral part of the production management data collection system. The "Cost Entry“ hypothesis is: Hypothesis 2: The events critical to effective CIM production management are different than the events that “trigger" cost account- ing data in traditional costing systems. Therefore, cost transaction triggers in CIM will be greater in number and diversity than those of the traditional setting. The ”Cost Entry” study area investigates this hypothesis. W When entered as transactions, costs are initially recorded in cost center accounts. Later, based on cost objectives, costs are assigned to products, activities, or functions. Two main purposes for cost assign- ment are to: (l) to match resource support costs with related resources and (2) to assign the total costs of resources consumed and used to production output.3 Some costs are easy to assign because they are easily identified with products or processes, such as materials or parts requisitioned. Assigning indirect support costs can be difficult if the costs cannot easily be related to products or processes (such as automated materials handling costs). CIM involves significant support costs such as main- tenance, logistics and other information, and technology-oriented kinds 3Indirect production support costs are described in more detail in Chapters 11 and III. Support costs make up a substantial portion of overhead costs in CIM plants. of production support [Cooper and Kaplan (1988); Miller and Vollmann (1985); Porter (1985 p. 67)]. Traditionally, direct labor has been used as the assignment factor for production support costs. However, the direct labor to overhead relationship is weaker in CIM because the set of resources in use is broader with greater use of non-labor, technological resources. Given the magnitude of non-material and non-labor costs as a proportion of total production costs, the cost assignment research hypothesis focuses on costs that have traditionally been pooled into the broad definition of overhead. Materials and labor are the only two classes of resources classi- fied as direct costs in traditional cost account groupings. Overhead pools, as traditionally classified, contain three kinds of costs: some direct production costs, many production support costs, and many facil- ities capacity costs. Since production support costs are substantial in CIM, cost management requires that support costs be assigned to the related class of resources supported (materials support assigned to materials resources, workers support to workers resources, etc.).4 To accomplish this, support costs in overhead pools can be reclassified into "resource support“ pools for each class of resource. Support costs can then be assigned to cost objectives based on factors that relate support costs with resources supported. Combining two purposes of cost assignment, the ”Cost Assignment" hypothesis is: “Figure 3 in Chapter II illustrates the patterns of cost assignment that exist when support costs are assigned by class of resources supported. 10 Hypothesis 3: Cost assignment in CIM will be changed by explicitly relating support costs with resource costs and assigning both types of cost to cost objectives by class of resource, using relevant cost assignment factors. This is contrasted with traditional manufacturing where all over- head costs are typically assigned based on direct labor or some other single factor. This hypothesis is investigated in the ”Cost Assignment” study area. W The final stage in cost accounting activities is to report costs, either in formal reports or by making cost data accessible by inquiry. Reporting requirements are derived from a variety of cost management needs, including strategic business management, production planning, manufacturing management, inventory determination, and financial report- ing. Cost objectives are derived from such various reporting require- ments. Reporting requirements are dictated in part by strategic manufac- turing objectives. While traditional manufacturing emphasizes produc- tion efficiency and resource utilization [Reeves and Turner (1972)], CIM's strategic manufacturing objectives are delivery reliability, quality assurance, production flexibility, and cost management [Buffa (1984)].5 The differences in emphasis of CIM's strategic manufacturing objec— tives affect cost reporting in several ways: 5Strategic manufacturing objectives are explained in greater detail in Chapter II. 11 - Cost data are used by resource managers more frequently in CIM and, therefore, cost data are processed more frequently. CIM operates in a fairly short time frame; many plans or actions are carried out within a daily or weekly time horizon. A monthly time frame is more typical in traditional manufacturing. m Due to inter-dependencies of integrated processes in CIM, costs are monitored at a more comprehensive, aggregate level. In traditional manufacturing, efficiency and cost effectiveness are monitored primarily at departmental levels. I CIM has a greater emphasis on cost management versus financial reporting. Cost reporting in CIM focuses primarily on strategic manufacturing objectives and secondarily on product costing for financial reporting, the opposite of traditional manufacturing. I Like cost entry, CIM cost reporting systems are likely to be an integral part of production data collection and reporting sys- tems. m Cost reporting in CIM is driven more by comparison with plans and future decisions than by past history, whereas traditional costing tends to focus on comparisons with the past. For all these reasons, cost reporting is expected to be different in CIM in terms of the form, format and frequency with which cost data are made available. Accordingly, the ”Cost Reporting" hypothesis is: Hypothesis 4: In the CIM setting, accounting information will be up- dated more frequently, “real-time” query form of reporting will occur more frequently, and the focus of periodic accounting reports will be on strategic manufacturing objectives versus the financial product costing objectives of the traditional setting. This hypothesis is investigated in the ”Cost Reporting" study area. Having described the role of cost accounting in connection with the research hypotheses, the terms "CIM" and "integrated processes" are now described in greater detail. 12 COMPUTER-INTEGRATED MANUFACTURING -- “CIM" Throughout the study, computer-integrated manufacturing (CIM) is defined as computer-aided technology applied to coordinated, interdepen- dent manufacturing processes. Literature defines "CIM" with a broader technical connotation. For example, Stover (1984, pp. 171-72) defines CIM as "a total process flow using computerized systems working with database management programs involving resource planning, engineering, quality control, flexible machining centers and materials handling systems.” Groover and Zimmers (1984, pp. 441-42) define CIM as "a complete integration of numeric control machines and robotics equipment with engineered process control, coordinated by computer driven mater- ials handling and process inspection systems." Goldhar and Jelinek (1985) describe CIM as "a combination of hardware, software, database management and communications technology for flexible automation.“ Descriptions of CIM in the literature emphasize both its technology and its integrative scope. CIM is more than state-of-the-art computer driven technology. It is the combination of advanced technology with comprehensive integration that makes CIM a radical change from tradi- tional manufacturing [Ettlie (1988)]. CIM may be placed in perspective by considering a continuum of automation. At one end is fixed (dedi- cated) automation; in the middle is flexible automated machinery in non- integrated, stand-alone configurations (sometimes described as ”islands of automation"); and at the other end of the continuum is CIM, inte- grated processes using flexible automation [Groover and Zimmers (1984)]. CIM links materials handling, engineering systems, production scheduling information, and integrated flexible process centers. CIM's scope of l3 integration includes: (1) business control systems with the production processes, (2) engineering design with manufacturing, and (3) the entire set of linkages between suppliers, production and key customers [Ettlie (1988, p. 11)]. Very few U.S. manufacturers fit the technical description of CIM of the preceding paragraphs. Most ”CIM" companies are in a transition from traditional manufacturing to CIM and do not have complete CIM systems in place, but have systems with various attributes of CIM. Therefore, for practical purposes for the study, CIM is defined in a more limited way as ”the combination of computer-aided technology and integrated manufac- turing processes." This definition seems consistent with the emphasis in the literature. Complementing this definition are certain CIM attributes that were considered in selecting the field research sites. These are described in Chapter IV (and listed there in Table 8): I Use of Materials Requirements Planning (MRP) in planning resource requirements I Use of manufacturing cells I Presence of "pull-through" control of production flow6 I Adoption of specific procedures to reduce set-up times I Use of computer-aided design (CAD) and computer-aided machinery (CAM) and robotics equipment. These attributes are consistent with the definition of CIM as computer- aided technology combined with integrated manufacturing processes. 6"Pull-through" control of production flow is usually associated with a 'just-in-time' (JIT) philosophy of delivering materials or inven- tory requirements just when required only in the quantities required. See Hall (1983) for further elaboration. 14 The definition of “integration” as used in the study is of more limited scope than that described in the literature. Integration as described in the literature includes (1) detailed data linkages between engineering design of components and products with specification of ma- chine processes, (2) planning and operational linkages between produc- tion and the total manufacturing data base system, and (3) functional linkages between purchasing, production and marketing [Gunn (1982); Vollmann, Berry and Whybark (1984)]. However, "integration" in the study is defined as and limited to linkages among manufacturing pro- cesses among manufacturing cells (product-focused clusters of flexible automation equipment) and within manufacturing cells. As explained by a production engineer at one of the field visit sites, CIM technology and information has made it possible to operate flexible processes together, like a ”symphony" of integrated resources rather than a "cacophony" of stand-alone dedicated process areas. Compared with other kinds of integration, the integration of manufactur- ing processes affects traditional patterns of cost incurrence and seems most relevant for the study. CONTRIBUTION OF THE STUDY The comparisons between traditional manufacturing and CIM provided by this study will help accountants and management assess the impact of introducing CIM into a traditional manufacturing environment. Accoun- tants have limited experience in assessing the impact of introducing CIM into a traditional manufacturing environment. The results of the study will suggest areas where research is warranted involving detailed 15 modeling and experimental discovery. Since the study findings are obtained from the context in which accounting is actually operating, the findings should help authenticate the theoretical basis of future research on the topic [Hopwood (1983)]. LIMITS OF THE STUDY This is an exploratory study that deals with an emerging issue. In general, CIM firms are in the early stages of reorganizing their pro- duction activities and developing their cost management systems. Revis- ions to cost accounting systems in CIM are still under development. Not many CIM firms exist for extensive comparison with traditional manufac- turing firms. Therefore, the study necessarily is drawn from a small set of firms whose revisions of their cost accounting systems are not well developed nor tested. The findings of the study reflect a process of change that is still in transition. Similar research carried out at another set of selected firms may not yield the same results obtained in this study. CONTENT OF OTHER CHAPTERS Chapter II establishes the context of the research problem by contrasting CIM with traditional manufacturing. Chapter II also ex- plains certain definitions and logical linkages to provide structure for the other chapters. Chapter III examines relevant background litera- ture, summarizes theoretical concepts, and specifies the study hypo- theses. In Chapter IV, the research design is specified, which is a field study involving visits to four traditional and four CIM 16 manufacturing plants. Chapter V analyzes the findings derived from empirical data gathered during visits to the field sites. Conclusions are summarized in two tables at the end of Chapter V. Results of the study and implications for further research are discussed in Chapter VI. CHAPTER II RESEARCH CONTEXT The purposes of this chapter are to describe the context of the research and to define certain terms and logical relationships used in later chapters. The chapter begins by comparing CIM with traditional manufacturing and describes the impact of CIM on cost objectives and cost accounting activities. After making these comparisons, the chapter then defines various terms and logical linkages for use throughout the study. MANUFACTURING: A MATTER OF RESOURCE MANAGEMENT To manage manufacturing activities is essentially a matter of managing resources: materials, workers, machinery and equipment, tool- ing, technology and information, and facilities. These resources pro- vide the capacity to convert procured inputs into saleable outputs. Between procurement of convertible resources and delivery of marketable products is a complex set of activities and technology that creates competitive advantage for a firm. Four resource management objectives are stressed in CIM literature: responsive flexibility, quality by design, delivery reliability, and cost management [Buffa (1984)]. W Prior to CIM, changing production conditions usually meant losing efficiency. Two patterns of competitive innovation, described by Aber- nathy (1976 and 1978) as ”fluid" and "specific,” are typically found in traditional manufacturing. The fluid state is the initial one, where 17 18 products with novel performance features are offered to customers. Costs and sales prices are both relatively high in this state. As unit prices are subsequently pushed down, the fluid state evolves to a speci- fic state of well-developed, standardized products [Abernathy (1978)]. Goldhar and Jelinek (1983) use the terms ”economies of scope“ and ”economies of scale" to describe the motive for the fluid and specific states, respectively. Because of enhanced flexibility, manufacturers can use advanced CIM technology to gain competitive advantage from both economies of scope and scale. "Flexibility" is defined as the capabil- ity of changing production output design or quantity in response to changes in market demand [Cohen and Zysman (1987, p. 131)]. Firms can use CIM technology to introduce new product designs quickly and effec- tively because costs of engineering design, testing and production inef- ficiencies during early stages of the product's life cycle are reduced [Gunn (1982)]. According to Cohen and Zysman (1987, ch. 9), flexibility in the CIM environment includes both ”static" and “dynamic” flexibility. ”Static" flexibility refers to near-term ability to change operations, whereas "dynamic" flexibility is the ability to respond steadily to changing technology by improving production processes and innovating product designs [Cohen and Zysman (1987, p. 131)]. Dynamic flexibility appears to be more cost effective in CIM than it has been in traditional manufacturing. According to Abernathy (1978), traditional manufacturers operate with a relatively low level of static flexibility, using standardized production processes in order to push down unit costs. Without standardization, manufacturing processes 19 were too inefficient and costly [Abernathy (1978)]. Traditional manu- facturers specify the engineering of their machine processes and workers' tasks carefully and introduce engineering changes cautiously and infrequently. According to Kusiak (1985), CIM can provide "systems flexibility,” enabling a firm to manufacture a variety of parts cost effectively, handle a variety of materials flow configurations, adapt its data sys- tems to changing specifications, and organize its activities to accom- modate near-term and longer-term changes responsively. The term "flexi- bility' is broad, encompassing systems for manufacturing, materials handling, and data base information, as well as organizational respons- iveness. 9111mm CIM technology permits consistent control of machine processes, which removes a significant amount of human error from manufacturing processes and improves the level of quality achieved [Goldhar and Jeli- nek (1983)]. Computer-aided technology helps to link product design with production processes, to set up processes correctly, and to detect when processes begin failing to meet process tolerances. Quality is "designed into" production processes rather than relying on ”inspecting out" quality defects [Sasaki and Hutchins (1984)]. In traditional manufacturing, quality is typically controlled by inspecting production samples according to acceptable levels of quality failures defined by policy. In CIM, however, sources of quality fail- ures are identified and eradicated rather than accepted [Sasaki and 20 Hutchins (1984)]. Quality failures arise in production when human effort lacks precision or when damage occurs as parts are packed, moved and unpacked. Integrated computer-aided processes help on both counts. Computer-aided processes are less dependent on imprecise human interven- tion, and less buffer work in process inventory is required between stages of production. Indeed, because advanced technology makes it feasible to have consistently high quality production, quality by design has become a source of competitive advantage. We: Delivery reliability is a vital manufacturing objective. Manufac- turers' customers demand delivery on time and in sequence so they can meet their own customers' delivery commitments. Managers want to retain flexible capabilities in the plant, produce to very high quality stan- dards, and yet, at the same time, supply product output precisely when needed. Delivery is reliable because production flow is controlled as it is pulled through process areas, rather than pu8hed through produc- tion to meet scheduled requirements. With "pull-through" control, new production of fabricated materials and components replenish consumed production in manufacturing process areas that feed other assembly. By contrast, traditional plants generally "push-through" production by releasing work orders based on anticipated production output require- ments. Managers forecast future requirements with the aid of scheduled production plans, inventory status records and Materials Requirements 21 Planning (MRP) systems.7 Orders for finished goods are filled from Open warehouse stock or back-logged for short periods until final assembly is completed. Management Cost management is a pervasive, underlying manufacturing objective that supports the other objectives. Traditional cost accounting meas- ures are generally viewed as accounting tools, used primarily to deter- mine product costs and secondarily to monitor cost control. In CIM, however, cost control takes on the broader perspective of cost manage- ment and is the primary, not the secondary, objective. Managing costs is a broader task than controlling costs. CIM firms compete with reli- able delivery of high-quality products, produced with responsive engi- neering design and flexible processes. In addition to achieving all of these objectives, they must manage costs. Firms that fail to manage costs may be at a serious competitive disadvantage [Cohen and Zysman (1987)]. Traditional manufacturers control costs by focusing on two primary production objectives: efficient production and capacity utilization. Materials efficiency is managed through scheduling and effective engi- neering; labor efficiency is managed with close supervision and cost ac- counting variance measures. In CIM, continued emphasis on cost measures that emphasize efficiency and utilization will not produce adequate 7"MRP" is defined in the APICS dictionary as ”a set of techniques which use bills of material, inventory data, and a master production schedule to calculate requirements for materials" [Wallace and Daugherty (1987, p. 18)]. See Orlicky (1975) for a detailed explanation of MRP. 22 information for cost management purposes. Efficiency and utilization alone do not address CIM management objectives. In fact, some writers have criticized traditional cost accounting for being too oriented to labor-based measures of efficiency and utilization, saying these meas- ures are not consistent with CIM's manufacturing objectives [Johannson (1984); Kaplan (1983); Wingard (1985)]. Since CIM's broad set of resources operate in an environment of integrated information and tech- nology, labor-oriented efficiency and utilization measures do not ade- quately capture cost implications of resource events. Wedge Reliability, quality, and flexibility have become the key strategic in CIM, ranking ahead of key traditional manufacturing objectives of stability, efficiency and utilization. Not that it is no longer impor- tant to stabilize the production environment, or use resources effi- ciently, or use available capacity -- it is a matter of strategic rank- ing of importance. The way to strategic, competitive success is to deliver high-quality products while having the capability to change quickly as the demands of the market change. The cost accounting chal- lenge is to adapt in ways that support cost management objectives. Strategic manufacturing objectives create greater demand for cost man- agement data than traditional cost systems have been accustomed to providing. 23 CONTRASTING TRADITIONAL MANUFACTURING AND CIM In general, CIM changes resource management decisions from reactive to proactive by providing managers the capacity to deal with complexity with reduced uncertainty. Table 1 compares CIM objectives with those of traditional manufacturing. 24 TABLE 1 COMPARISON OF MANUFACTURING OBJECTIVES Objectives of the M132— Responsive flexibility: With in- tegrated, coordinated processes, production can accommodate variety in quantity and mix. New product designs are introduced quickly and efficiently. Firms can respond effectively to short-term changes in market demand. Economies of scope outweigh economies of scale. Quality by design: Computer-aided processes are precise without depending on human skills. Qual- ity is "designed into” production processes rather than depending on ”inspecting out” quality failures. Prevention of quality failure is the key. Delivery reliability: Delivery reliability is a competitive objective. Final assembly is scheduled with very high reliabil- ity; other production is "pulled through" process areas to supply final assembly. Cost management: Costs of CIM's variety of conversion resources, including technology and informa- tion, are managed rather than merely controlled. Cost manage- ment systems focus on manufactur- ing objectives of flexibility, quality, and delivery reliability, and measure events involving all cost drivers, including technology and information. The aim is to minimize integrated production costs at the macro level. Objectives of the W Stability: Economies of scale are the key to cost competitiveness, with reliance on stabilized pro- duction to drive costs down pre- cludes much variety in product designs. Processes are sequential and not well linked. Engineering design changes are introduced cautiously and infrequently. Quality by inspection: Quality is typically controlled by sampling and inspecting production accord- ing to acceptable levels of qual- ity failures defined by policy. Detection of quality failure is the key. scheduling of released production: Production of complex, assembled products are planned with Mater- ials Requirements Planning (MRP) systems. Scheduling focuses on release of work to the shop floor. Production is subject to "push through” control. Cost control: Traditional manu- facturers provide information on cost control, focusing on two production objectives: efficiency in the use of materials and labor, and utilization of labor and machine capacity. Cost minimiza- tion is aimed at the migrg level. 25 The contrasts between CIM and traditional manufacturing relate mostly to the additional capabilities that computer-aided technology can provide and improved linkages between highly integrated processes. Because of the power of technology and information in CIM, certain long- time assumptions -- such as the need for a high degree of stability, buffer inventories, and a cost-effective quality failure rate -- are replaced by others more suitable. The new assumptions have important implications for cost accounting activities. IMPLICATIONS FOR COST ACCOUNTING ACTIVITIES Flexible production, with consistent high quality, inventory mini- mization, and pull-through production control -- these are dramatic changes in the cost accounting's implicit model of manufacturing activ- ities. Failure to adapt cost accounting systems to reflect a different manufacturing model may inhibit managers' efforts to manage costs in CIM environments. For example, managers may not be able to determine the potential impact of cost reductions on total product costs, or engineers may not be able estimate accurately the effects of proposed engineering changes. Changes in cost accounting systems are not made easily. Cost ac- counting measures are relied upon to provide data for financial state- ment reporting. Two legitimate concerns are that a transaction audit trail be maintained and that transactions be recorded consistently between financial reporting periods. These concerns explain a degree of reluctance to change cost accounting practices. CIM's organizational and technological characteristics suggest changes in cost accounting 26 activities, including revised cost classifications for identifying costs (groupings of accounts according to resources and redesignation of cost centers), changed data collection techniques, revised assignment of resource support costs, and changed ways of reporting cost data. WW Categorization includes specifying names of accounts, grouping of accounts for cost assignment purposes, and designating boundaries of cost centers. While specifying names of accounts is a matter that depends on each particular firm, classification schema presumably will change to better reflect the kinds of resources used in CIM. grgup1ngg_gf_ggggunt§_by_re§ggrge§. A traditional practice has been to group costs into two "direct" classifications (direct materials and direct labor) and an "indirect” classification (overhead). Mater- ials and labor have been considered "direct” costs because materials quantities and labor hours can be identified with units of output. In addition to materials and labor, other resources are used in production, such as machinery and equipment, tooling, and facilities. Since in traditional manufacturing these resources cannot easily be identified with units of output, the costs are classified as overhead and allocated to cost objectives, usually in proportion to direct labor. In effect, direct labor is used as a surrogate for costs of other resources. Direct labor is not a satisfactory surrogate for technological resources in CIM. Compared with traditional manufacturing, a smaller proportion of total labor is direct. CIM workers handle a variety of duties and work in teams rather than as individuals. Their time is not 27 concentrated on specialized, narrowly defined tasks. Therefore, since output cannot be associated with individual workers, much of their time is classified as indirect labor. With more labor being classified as indirect and with increased costs of technology recorded in overhead, a higher amount of overhead is applied with a reduced amount of direct labor. In addition, direct labor is a less appropriate surrogate for costs of non-labor resources because the cost incurrence patterns of the other resources are poorly reflected by the cost incurrence pattern of direct labor. Relying on direct labor to assign costs of technological resour- ces can have the effect of obscuring the transaction events that drive the costs [Cooper and Kaplan (1988)]. But the traditional approach is to pool costs of technological resources and redistribute them to cost objectives based on direct labor. If this is no longer appropriate, what is the alternative? The alternative is to expand the set of resources identified as direct resources and to group remaining indirect costs into more homo- geneous pools. Traditional cost systems identify materials and labor as the only direct resources, with all others being considered "overhead." Replace these classifications in CIM with a wider set of resource clas- sifications. Based on observations made in the field during this study, a recommended set of resource classes is: materials, workers, machinery and equipment, tooling, technology and information, and facilities.8 Other classification sets can be suggested, but this particular set 8These classifications may help to resolve problems of cost distor- tions arising in ”two-stage” allocations of overhead costs as described in Johnson and Kaplan (1987, ch. 8), and in Cooper and Kaplan (1988). 28 seems consistent with the field research observations and will be used in the study as a common reference point. Each of these classes of resources have separate instances of use, are more or less documented separately in transactions, and are controlled separately. These resource classifications can be adopted in cost accounting systems in place of the traditional classifications of direct materials, direct labor, and overhead. Direct costs can be assigned from each resource class to cost objectives (products, cells or cost centers) based on appropriate measures of ”cost driver" events that require the consump- tion or use of the resources.9 Overhead costs in CIM plants include not only technological resource costs but also substantial costs of supporting resources, which can be classified by class of resource. An overhead cost pool can be sub-divided into separate pools of support costs for each class of resource. These support pools can be assigned to direct resource cost groups based on resource transaction factors, such as those suggested in Miller and Vollmann (1985). Then the combined costs of resources and support can be assigned to cost objectives with more authentic reflec- tion of cost drivers than possible with traditional resource account groupings. Table 2 compares account groupings for CIM resource costs and related support costs pools with traditional cost accounting groupings. 9See Porter (1985), chapter 3 for additional discussion about cost drivers reflected by transactions. 29 TABLE 2 COMPARISON OF RESOURCE ACCOUNT GROUPINGS RESOURCE ACCOUNT GROUPINGS IN CIM W Materials Direct materials Workers Direct labor Machinery and Equip. Depreciation Operating costs Tooling Tools, jigs and fixtures Technology and Information Data systems Software Engineering Facilities Building deprec. Building maintenance Maintenance and repair Insurance W Materials Support Indirect materials Materials handling Stockroom costs Workers support Indirect labor Supervision Employee benefits Training Mach. & Equip. Support Repair and Maint. Set-ups Tooling Support Tooling design and Repair Technology and Information Support Amortization Maintenance Facilities Support Custodial services Utilities TRADITIONAL GROUPINGS W Materials Direct materials Workers Direct labor Manufacturing Overhead Indirect materials and materials support Indirect labor and workers support Machinery & Equipment and support Tooling and support Technology and information Facilities and support By grouping accounts into resource classes, production support costs can be planned and controlled in connection with resources being supported. This is not feasible when several resources and their sup- port costs are co-mingled in a single grouping of manufacturing overhead 30 costs. Furthermore, the term ”support" has a more appropriate connota- tion to resource managers who are demanding or providing support. Dgg1gn§§12n_2£_29§§_§gn§g1§. Traditional cost center boundaries generally match specialized process areas in manufacturing, such as fabrication, machining, welding, painting or finishing areas. Process specialization distinctions are blurred in CIM, where resources are organized to focus on products rather than processes. Product-focused CIM cells typically include a variety of processes. Cost centers in CIM encompass broader manufacturing capabilities, but are smaller and more numerous than the larger specialized cost centers of traditional plants. Where production flow is controlled by ”pull-through" based on scheduled delivery of finished products, fabricating areas produce on a replenish- ment basis. Figure 1 illustrates the effects of pull-through produc- tion flow by comparing the floor layouts of CIM and traditional plants. Receiving Receiving Raw Materials T Raw Materials I Finished Fabricating Areas Fabricating Area Goods Cells for product families \ / \ / \ / Warehouse \ / \ / \ / I I I I I I Assembly Area Area Finished Goods Warehouse Area Shipping Shipping Dock Dock Floor layout -- CIM Plant Floor layout -- Traditional Plant (Pull-through production flow) (Push-through production flow) Figure 1: Cbmparison of Floor Layouts, CIM and Traditional Plants 31 In the CIM plant, production flow is in a pull-through direction (bottom of the figure to the top), with the impetus of the ”pull" origi- nating from scheduled deliveries out of the Finished Goods Warehouse Area. These scheduled deliveries are supplied by product family cells, where virtually all of the conversion of the product occurs. Cells draw fabricated materials from adjacent containers filled by the Fabricating Areas, where production is initiated only when containers become empty. That is, production is not scheduled in Fabricating as in the cells. Instead, replenishment is activated when a need is made apparent by an empty container. Production in the traditional plant begins with the release of scheduled orders into the Fabricating Area. These orders are committed into assembly in sufficient time to meet a scheduled delivery commitment into the Finished Goods Warehouse Area. Production flow 'zig-zags' from top to bottom in the figure. The smaller areas for Raw Materials and Finished Goods in the CIM plant reflect the inventory reduction benefits of pull-through 'just-in-time" production control [see Hall (1983)]. Furthermore, the wider shipping area in the CIM plant allows products to be delivered from cells very soon after production is complete. The changes in plant lay-out and direction of production flow are dramatic. Pull-through production flow improves coordination and per- mits more production throughput in a smaller area of floor space. Replenishment-oriented areas require less tracking by inventory control systems. These changes affect data collection procedures as different resource events are chosen as transaction triggers. 32 W In replenishment oriented fabrication areas of a pull-through CIM plant, less inventory tracking is needed than in traditional plants. Programmable microprocessor controllers provide the technical specifica- tions to the fabricating equipment so that the same equipment can make a variety of parts. Standardized containers filled with pre-set quan- tities make it easy to see the inventory already produced for the cells. In short, visual information is sufficient for operators to know their inventory status. Cell areas have new trigger points suitable for flexible produc- tion. Included are points of cell entry or exit, which can be used to measure elapsed time. Some resource costs and support costs could be assigned based on elapsed cell time. Other points where conformance with quality or other specifications is affirmed may provide triggers to measure quality assurance costs or to detect and record waste. CIM has technology that can facilitate effective data collection techniques. Electronic reading of magnetic strips, or “bar coding“ is an example. With this technology, it is feasible to choose appropriate resource events for transaction triggers. In certain areas, more trig- ger events may be needed to satisfy requirements for up-to-date informa- tion about production status if not obvious from visual information. Also, CIM requires more detailed evidence of the incidence of costs, and this can be provided by appropriate triggers. The opportunity for choice of triggers is more limited in traditional manufacturing, where cost transactions are typically triggered when production is initiated or completed (entry and exit), or transferred between inventory storage 33 locations. Figure 2 compares trigger points in CIM with traditional manufacturing. Receiving Receiving Raw Materials } Raw Materials ~—[tl [t1 [t] I t-[t] [t] Finished Fabricating Areas Fabricating Area +——[t] [t] [t] Goods Cells for product families \ [t]/ \ / \ / r—[t] Warehouse [t]\ /[t] \ / \ I I I I I I Assembly Area Area It] [Cl—[tl— [1:] Finished Goods Warehouse Area I ‘ Shipping Shipping Dock [ Dock —[tl———-[tl-—-—[tl [cl—— Trigger points -- CIM Plant Trigger Points -- Traditional Plant '[tl' symbolizes places where resource events are trigger points. Figure 2: Comparison of Transaction Trigger Points In the CIM plant, trigger points of entry into each of the fabrica- ting areas indicate where requisition withdrawals from the Raw Materials Stockroom occur when the Fabricating Areas replenish their supply. The act of replenishment is initiated when a cell team empties a standard container of fabricated parts, which signals a need to refill it. The replenishment act is quite similar to a reimbursement of a petty cash fund. No trigger is needed when the container is filled; instead, the inventory on hand in the containers (which contain standard quantities) is counted when necessary at the end of accounting periods. Trigger points exist at entry and exit from the cells, so that elapsed time can be measured. Note that because of advanced data 34 collection capabilities, multiple triggers within cells are available as needed to measure conformance with specifications, to track production status, or to collect cost data. Figure 2 illustrates multiple trigger points in the first cell. Additional trigger points occur at shipment. Trigger points in the traditional plant are where raw materials and components are requisitioned into production, production leaves the Fabrication Area to go into the Assembly Area, and finished product leaves Assembly to go into the Finished Goods Warehouse Area. Payrolls and shipments from the warehouse provide additional triggers. Shipment is made directly from the area near the cells of the CIM plant. WWW}. Production flow in traditional manufacturing depends largely on direct laborers' efforts. However, production flow in CIM relies exten- sively on information-oriented activities, such as ordering, scheduling, releasing, receiving, and transferring. These activities are vital for effective production management, but are only indirectly associated with the line worker efforts. The proportion of manufacturing input repre- sented by direct labor effort decreases in CIM, while the proportion represented by information, technology and other indirect costs in- creases [Miller and Vollmann (1985)]. Accordingly, traditional overhead cost pools are likely to be divided in CIM by regrouping some costs into direct resource pools and by subdividing support costs according to the groups of direct resources. The objectives of such a classification are to match support costs with relevant resources and to assign support 3S costs to cost objectives consistent with the assignment of the resources costs. Figure 3 illustrates the arrangement of resource groupings with re- lated support pools. Materials Workers Mach. and Tooling Technology Facilities Support Support Equipment Support and Infor. Support Support Support 8 U u u 8 8 Materials Machinery Technology and Workers and Tooling and Facilities Components Equipment Information L——l_:_ 8 U 8 fl 8 U u U U u I u Cost Obj. 1 Cost Obj. 2 o . . . . . . COST OBJECTIVES (A job, cell, or other cost center) Figure 3: Patterns of cost Assignment In this figure the resource costs are grouped into six cost pools, with cost assignments of resource-related support costs to those groups initially, followed by assignment of combined resource and support costs to cost objectives. Although only two cost objectives (products or cost centers) are illustrated, the assignment pattern would apply just as 36 well to any cost objectives (which are listed in the next section in Table 3). Moreover, Figure 3 is not meant to imply that only those six groups might be established in any particular firm. Some firms may decide to combine some of the groups into less than six; others may find more than six desirable. It is not the number or even the names of the groups that is most important, but rather the notion that resources should be grouped in such a way that more costs can be accounted for as direct costs and that cost assignment can be improved. Figure 3 follows the sequence of assigning support costs first to relevant resource groups and then assigning combined resource and resource support costs to a cost objective. For any of the resource groupings, there could be other variations in the cost assignment pat- tern. For some resource groupings in certain circumstances, it may be appropriate to establish a combined pool of resource costs and support costs and use a single assignment factor for the combined pool. In other cases, there may be separate pools for the resource and support costs but with a single cost assignment factor and separate assignment. Or, there may be separate pools and separate (different) assignment factors. These alternative cost assignment patterns are illustrated in Figure 4. 37 Technology Workers Machinery Tooling and Infor. Support and Equip. Support Support Support Technology I ° and O O O O O O O 0 0 Information 3 ' ' Machinery ‘ ° Workers and ' Tooling ‘ Equipment ' ' O - O 0 . 0 , -0 - I 0 0 O O O 0 - O —| ° I I I I I I I I I I I ° — — _e_.__| I . l O - I O O 9 O O 0 0 ll ll ll I! It It 8 I] I ll ll 8 Cost Obj. 1 I Cost Obj. 2 o o o - - COST OBJECTIVES (A job, a cell or some other cost center) Figure 4: Alternative Patterns of Cbst Assignment Figure 4 illustrates four alternative cost assignment patterns: The first support pool, Technology and Information Support, is classified in the same pool as the related resource costs and assigned with one factor as combined costs. Support for the Workers resource group is pooled separately, but then assigned to the Workers resource cost group before assign- ment together to cost objectives as a combined cost. The third resource, Machinery and Equipment, has a separate sup- port pool. Both the resource costs and support costs are as- signed separately, but with the same assignment factor. The fourth resource, Tooling, not only has a separate support pool, but is assigned separately with a different factor for support costs (e.g. number of set-ups or hours of set-up time) and for resource costs (e.g. cost of tools actually issued or installed). The Tooling support and Tooling resource assignment factors are symbolized by ”" and "a,” respectively. 38 The key to assigning costs to cost objectives is to identify and use cost driver factors that measure resource use or consumption. Traditional cost assignment methods may combine many heterogeneous costs and therefore poorly reflect cost drivers. WWW Cost reporting requirements are driven by cost objectives. Cost reporting requirements are different in CIM than in traditional manufac- turing. Traditional cost accounting reporting is based on stabilized data sources and periodic reporting cycles. CIM has more variety and more frequent change; cost data is accessed more frequently. CIM's integrated processes mean costs are monitored at a more comprehensive level, cost data must be more timely, and up-dated more frequently. CIM has a daily or weekly time horizon, compared with traditional manufac- turing where a monthly time horizon is typical. Cost objectives serve more purposes in CIM. Costs are used to help manage resource demands. Evaluation of integrated process operations is aided by cost data. There is constant monitoring of quality failures and other kinds of costly production failures. By contrast, cost objec- tives in traditional manufacturing relate primarily to tracking costs to cost centers, measures of efficiency and utilization, and product cost- ing for inventory determination. The contrast between cost objectives in CIM and traditional manufacturing is summarized in Table 3. 39 TABLE 3 COMPARISON OF COST OBJECTIVES CIM COST OBJECTIVES TRADITIONAL COST OBJECTIVES Strategic Manuf. Objectives Inventory cost determination Delivery reliability costs (proper loading of Flexibility costs inventoriable costs) Quality assurance costs Resource needs and status Capacity utilization Macro cost control at Micro cost control at integrated levels cost center levels Production planning Labor productivity Efficiency Failure costs Utilization Emergency maintenance Set-ups Budget preparation Quality Variances Resource waste costs Actual vs. plan Actual vs. standards Budget preparation Variances Actual vs. plan Actual vs. standards Inventory cost determination (Proper loading of inventoriable costs) More cost objectives exist in CIM, reflecting the use of cost data for cost management. Costs are needed both at the level of local cells or work stations and also at higher integrated levels. Detection and elimination of failures is a point of emphasis in CIM, including such failures as emergency maintenance, avoidable set-ups, or quality fail- ures. Cost objectives exist for each of the strategic manufacturing objectives of delivery reliability, flexibility, and quality. 40 Wired Table 4 compares the cost accounting activities of each area: TABLE 4 COMPARISON OF COST ACCOUNTING ACTIVITIES Cost Accounting Activities in a CIM Environment Cost Identification: Process areas are organized around product fami- lies. Clustered resources in cells means smaller, more numerous cost centers. Cost accounts for resour- ces and related support are grouped in multiple resource classes, in- cluding machinery, tooling, tech- nology and information. Cost Entry: More resource events are available for transaction trig- gers. Trigger events are where pro- duction enters or exits a process area, or where conformance with specifications is affirmed. Advanced data collection technology (i.e., bar coding) makes more triggers feasible. Cost Assignment: Costs of tech- nology, information and other in- direct support are a larger pro- portion of total product costs. Redefined overhead cost pools relate resources with resource support costs resulting in improved use of relevant cost assignment factors. Cost Reporting: More cost objec- tives exist to report cost manage- ment data, including measures of costs incurred to meet plantwide objectives of flexibility, quality, and delivery reliability. Cost data are updated frequently, kept timely, and accessed frequently; the cost data time frame is daily or weekly. Cost reporting is an integral part of production management reporting. cost Accounting Activities in a Traditional Environment Cost Identification: Cost center boundaries generally match those of specialized areas in manufac- turing, such as fabrication, machining, welding, painting or finishing. Resource groupings are materials, labor and over- head. Cost Entry: Cost transactions are typically triggered when production is initiated, com- pleted (entry and exit) or trans- ferred between inventory storage locations. Cost Assignment: A substantial portion of overhead costs are presumed to be labor-related. Overhead costs are commonly ap- plied on the basis of direct labor. Cost Reporting: Cost accounting systems rely on stabilized budget and standard cost data files and follow a monthly reporting cycle aimed predominantly at product cost for financial reporting. Cost reporting is separate and often independent of production reporting. 41 As indicated by Table 4, cost accounting's role is changed by CIM in each of the four activity areas. Account groupings and cost center designations will be changed because the organization pattern of resources and production flow patterns will be changed dramatically. There are likely to be more transaction triggers so that cost data can be obtained frequently and can be aggregated to meet a need for macro level data that measures aggregated plant-wide costs. Cost assignment will change to reflect a revised configuration of resources and resource support pools. There are a greater number of cost objectives to be satisfied by reported costs. W At this point it is appropriate to summarize several of the defined terms mentioned throughout Chapters I and 11. Manufacturing is essentially a matter of managing resources with distinct activities intended to add or maintain production value; these activities are cost driver activities [Porter (1985)]. Costs are meas- ures of resource events that result from resource management decisions and cost driver activities intended to accomplish strategic manufactur- ing objectives (flexibility, quality by design, delivery reliability and cost management). A resource event involves either consumption of an expendable resource or use of a durable resource. Resource events may contribute directly to the conversion of particular resources into other resources, or may support conversion directly or indirectly. Cost identification, cost entry, cost assignment and cost reporting are the four activities that comprise the role of cost accounting. The 42 cost identification and cost entry activities recognize, classify and record costs based on the nature of the resource events require consump- tion or use of resources. Cost assignment aims at linking the recorded costs with cost objectives, which are the bases for cost reporting. A cost objective is any purpose for which a cost is reported, including cost estimation, inventory determination, or a variety of cost manage- ment purposes, such as labor efficiency, capacity utilization, or qual- ity failure monitoring. Cost objectives are driven by production man- agement, marketing and financial reporting requirements. SUMMARY This chapter has contrasted objectives of CIM with traditional manufacturing and has described four activities that comprise the role of cost accounting. Expectations regarding the impact of CIM on cost accounting activities and objectives are stated. In the next chapter, support for these expectations is developed from background literature, concluding with the research hypotheses. CHAPTER III BACKGROUND FROM RELEVANT LITERATURE INTRODUCTION The role of cost accounting is defined in Chapter I as four activities: cost identification, cost entry, cost assignment and cost reporting. Chapter II described how CIM has changed the context in which the role of cost accounting is carried out. This chapter sum- marizes background literature relevant to the role of cost accounting and the impact of CIM on that role. Two classes of literature are examined: (1) literature on traditional cost accounting concepts and practices and (2) literature on the nature of CIM and its impact on cost accounting. Literature on traditional cost accounting is important for under- standing traditional objectives of cost accounting in the historic context from which those objectives originated. Part of the process of adapting the role of cost accounting is to reinterpret these traditional objectives in the light of contemporary cost management objectives. Literature on CIM establishes the characteristics of CIM and suggests the implications for cost systems and the role of cost accounting. The background literature was reviewed with a particular aim in mind: to identify theoretical points that could serve as a framework for developing the theory for this study. Theoretical points discussed throughout the chapter are the basis for the research hypotheses speci- fied later in the chapter as well as several "focus points" specified for each hypothesis. These focus points are used later in Chapter V to organize the discussion of the empirical findings. 43 44 LITERATURE ON TRADITIONAL COST ACCOUNTING Today's cost accounting systems include concepts and practices traceable to historic origins of traditional manufacturing [Kaplan (1984a)]. The literature reviewed in this section establishes how some basic elements of today's cost accounting systems first emerged. The section begins by defining and describing the term ”traditional manufac- turing" for purposes of the study. Then, historic literature on cost accounting is examined. This historic literature reveals the origins of concepts and practices still existing in today's cost systems, including labor-oriented costing, use of standard costs, assignment of manufac- turing burden (overhead), cost centers, and the notion of “attaching” costs to products. Then early literature on standard costs is reviewed because it provides additional conceptual insights into the intended role of cost accounting when traditional manufacturing was developing. In the final part of this section, a field study of the role of the con- troller is examined because its findings help to explain cost accounting activities. TRADITIONAL MANUFACTURING DEFINED For purposes of the study, the term “traditional manufacturing” is defined as ”make-to-stock,” batch-oriented conversion of materials and components into discrete products built for inventory for later sale [Harrington (1984), Reeves and Turner (1972)]. There can be variations of this description in practice. For example, some traditional manufac- turers are organized as ”make to stock" for components but "assemble to 45 order" for final products.10 For more specialized production, other traditional manufacturers produce on a ”make to order" or, in some cases, "engineer to order" for highly specialized products. The key distinction here is that I"traditional manufacturing" is used in this study to describe relatively large volume production with stabilized, dedicated processes. Traditional manufacturing has had the greatest influence on the development of existing cost accounting concepts and practices [Chandler (1977, ch. 8,14); Kaplan (1983 and l984a)]. Since the early part of this century, manufacturers have sought economies of scale by organizing production with dedicated machine processes and fixed assembly lines [Chandler (1977), Hayes and Schmenner (1978), and Cohen and Zysman (1987)]. Stability is a key element in traditional manufacturing strategy. Product designs for new products are tailored to process requirements, thus enabling production volume to rise quickly and realize economies of scale [Abernathy (1978)]. Uneco- nomic production was often attributed to mismatching of product charac- teristics with process characteristics [Hayes and Wheelwright (1984, ch. 7)]. Traditional manufacturing was viewed primarily in operational (not strategic) terms, where the emphasis was on stability, efficiency, and capacity utilization. In traditional manufacturing, workers were rewarded for their skills and rapid task performance; they worked as individuals. In summary, two attributes of traditional manufacturing are sig- nificant for the theory of this study: 10For further discussion of make-to-stock and make-to-order produc- tion, see Buffa (1984, ch. 3). 46 - Historically, traditional manufacturing has been viewed as a costly activity but not a source of competitive strategy. Cost effectiveness was a paramount focus of management attention, which has led to an emphasis on measures of labor efficiency and labor utilization. - The worker in traditional manufacturing has generally been viewed as the key to attaining production efficiency and effec- tiveness. The worker added specialized skill to generalized technology. Understandably, cost systems emerged which provided detailed data to management about the workers' performance, with far less detail about other resources used in production. HISTORIC ORIGINS OF TRADITIONAL COST ACCOUNTING Existing cost accounting concepts and practices originated from the context of traditional manufacturing. Literature on the historic orig- ins of cost accounting establishes the historic basis for labor-oriented costing, standard costs, manufacturing overhead, cost centers and cost attaching to products.11 Product costing can be traced back to "job shop” production of the latter part of the nineteenth century. Foremen of that era functioned as “inside contractor” employees and were compensated as independent contractors [Chandler (1977, p. 271) and Litterer (1961b)]. After sub- stantial growth in the scale of manufacturing early in the twentieth century, the role of the foreman began to change to that of a supervisor specialist, subject to more detailed accountability for production costs. When electric power and transportation became more widely avail- able, manufacturing firms expanded and their production operations became more diverse. Complex, dedicated process technology came into 11For additional discussion of the historic origins of existing cost accounting concepts and practices, see Kaplan (1984a), Chandler (1977), and Litterer (1963). 47 use for mass production of standardized products. Knowledge of this technology resided primarily with the foremen. Top management began to rely on cost accounting data to control the decisions and actions of foremen. Cost accounting systems became formalized and refined [Chandler (1977)]. MW This was the period of the emergence of ”scientific management” promoted by production engineers and attributed to Taylor (1911). During that same period, some engineers (including Alexander Hamilton Church, Harrington Emerson, and Henry R. Towne) published articles on organizing "systematic management” cost systems compatible with scien- tific management [Litterer (1963)]. This early literature advocated categorization of direct and indirect costs and assignment of overhead costs based on direct labor [Chandler (1977, Chapter 8)]. Engineering data systems, developed between 1910 and 1920, made it possible to track costs as a means of evaluating process control and productivity. Formal "scientific management” procedures developed by Frederick W. Taylor became the basis for cost accounting measures of dissimilar operations in multi-divisional manufacturing companies [Taylor (1911)]. The en- gineering data were labor-oriented, so cost accounting became labor- oriented. W The rudiments of today's cost accounting systems were formalized during the historic period when mass production emerged. The worker was 48 a principal resource used in conjunction with complex process technology to produce in large volumes. Cost measurement systems were developed to help upper management control production without detailed supervision. Developments in engineering data systems also helped to formalize stan- dard cost accounting systems. Engineers used "scientific management" concepts and procedures to standardize production tasks and time re- quirements and save costs. Production laborers' tasks were formalized and specialized through job analyses and time and motion studies devel- oped by Taylor, Gantt and others. [Chandler (1977); Taylor (1911)]. W Specialization led to departmentalizing of production activities and the creation of factory staff support positions, including time- keepers and production control clerks. With the creation of such posi- tions, production control became a costly and necessary organizational burden to ensure control. The support costs became known as "factory burden," which were assigned from cost centers to products, so that all supportive costs were allocated to production as it flowed through the plant [Chandler (1977, p. 278)]. W Alexander Hamilton Church advocated cost centers to provide control of overhead costs [Litterer (l961a)]12. Church defined five 12Based on publications by A. Hamilton Church cited in Litterer (l961a): "Practical Principles of Rational Management," Enginggzing Magazing, Vols. 21,22 (1901); later published in: A. Hamilton Church, £19§ng§19n_£ng§gzg. New York: Engineering Magazine Co., 1910. 49 manufacturing functions -- Design, Equipment, Control, Comparison and Operation -- and developed cost center measures for each. Early efforts to formalize cost accounting in factories were furthered in Clark (1923), a comprehensive examination of the nature of overhead costs, and by a very detailed specification of standard costs systems in Harrison (1921 and 1930). Church, Clark and Harrison helped document cost ac- counting concepts and practices applicable to that era of growth in manufacturing and scientific management. Much of what they wrote is still found in today's cost accounting practices [Chandler (1977); Johnson and Kaplan (1987); Kaplan (1983, 1984a)]. A strong connection between direct labor and manufacturing overhead developed during this era. There were two sources of overhead costs, support for workers and production accounting costs, both of which were considered legitimate production costs. Factory burden was justified as a means of management supervision and control, made necessary by the specialized nature of worker tasks. werkers' skills were considered a resource to be used carefully. Costs of supporting workers required accountability in factory burden accounts. Since the worker was the primary resource focused on by management, allocation algorithms based on direct labor were used to assign factory burden to product costs. AWE Traceable back to early cost accounting systems is a notion that costs can be "attached" to products as they travel from functional departments [Johnson (1987), Johnson and Kaplan (1987, p. 132)]. As in- dustrial engineers organized production in a smooth flow through 50 functional departments, engineered standards began to be used in cost accounting measures. Engineered bills of materials were used for meas- ures of required materials, operations routings data were used for meas- ures of direct labor hours, and labor time was the basis for assigning other indirect costs from burden pools [Chandler (1977, ch. 14)]. Mean- while, auditors found engineered data to be an attractive and objective means of assigning costs to inventory. Thus, it became common practice to attach costs to products as materials flowed through various stages of production. Cost attaching procedures eventually became the basis for accepted cost accounting concepts [Johnson and Kaplan (1987)]. Originally, cost accounting helped managers control costs; cost data instilled cost accountability. But as accountants and auditors recognized the convenience of using cost systems to determine inventory by ”attaching" costs to products and as the influence of financial re- porting became stronger, cost attaching began to out-rank cost managing as the focus of cost accounting [Johnson (1987)]. To summarize, early cost accounting systems used formalized en- gineering specifications of materials and labor to attach costs to products and determine inventory costs. In addition, engineering data provided the means for development of standard cost systems. The meas- ures were labor-oriented because the focus of control was primarily on the worker's tasks. The next major source of development of cost con- cepts and practice was the development of standard cost systems. 51 STANDARD COST SYSTEMS A "standard cost" is a target measure of expected production cost under normal manufacturing circumstances.13 Standard costs are calcu- lated for a product or component by multiplying normal quantities of materials, component parts and labor hours times normal prices and wage rates. Concepts developed for standard cost systems are relevant for theory about cost accounting. Since many standard cost concepts have become accepted as cost accounting concepts, standard cost literature were examined in search of further theoretical support for study. Out of the body of standard cost literature, Harrison (1921 and 1930), Henrici (1960), N.A.A. (1974), and Dewelt (1975) were selected as repre- sentative sources. W G- Charter Harrison's WW (1921) and WW (1930) describe standard cost concepts and methods developed by engineers from scientific manage- ment theory. Harrison (1921) includes the following areas of emphasis pertinent to the present study: a Standard costs should specify expected costs both in total and by detailed operation, pre-determined under normal operating conditions (pp. 7-8). m Standard costs should distinguish between productive costs and idle costs (p. 9). 13N.A.A.'s terminology publication (1983, p. 100) defines standard cost as “a forecast of the cost of performance that should be attained under projected conditions as determined by reasonable estimates or engineering studies." 52 m The accounting function should establish the standard costs, based on data determined and maintained by engineering (p. 36). m Information reports comparing standard and actual costs should be prompt and accessible to production managers (p. 12). m Cost reports should focus on exceptions that require management attention (p. 14). These areas of emphasis are continued with greater elaboration in Har- rison (1930), which includes several detailed charts showing the flow of transactions through the standard cost system. Several conceptual ideas are derived from Harrison's views on stan- dard costs. Standard costs help reveal where costs are incurred in processes involving inter-related resource events and should help mana- gers isolate wasted costs from productive costs. Standard cost measures should be coordinated with measures used by engineering and production. Coordination keeps cost data current and accurate. Coordination helps managers understand and accept that cost data are measurements of pro— duction results rather than merely accounting results. This encourages managers to feel a sense of ownership of cost data. While accountants are responsible for and "own" cost measures (measuring tools), manu- facturing managers are responsible for and "own" cost measurements (measured cost data). Managers can easily perceive accounting measure- ments as "accounting's numbers" rather than a reflection of their own resource accountability. Harrison's works are an important source of cost accounting con- cepts. They articulate the motives and objectives of cost data in an era when cost identification, entry, assignment and reporting practices were becoming widely accepted, many of which have lasted to the present day. 53 W Henrici (1960, pp. 31-32) defines "cost center” as a ”unit of endeavor under the lowest level of supervision, buying materials and services from other centers, incurring expenses within itself, and in turn perhaps selling materials and services to other centers." Henrici said costs should be measured at the level of process operations and defined an ”operation" as "a plant activity at the first degree of subdivision, which has a known unit of output and whose costs differ from those of other activities" [Henrici (1960, p.33)]. Operations identified in standard costs should be coordinated with the operations identified for control of manufacturing processes. With coordinated definitions, costs of individual operations can be aggregated to measure total cost of production. Henrici also warned against combining costs of separate cost centers, thus concealing traceability. WE). DeWelt, an accountant employed by a heavy equipment manufacturer, wrote about using standard costs as an aid for inventory control where Materials Requirements Planning (MRP) is used.14 DeWelt explains why standard cost data and reports need to be timely and coordinated with MRP processing cycles. Since MRP focuses attention on scheduled pro- duction and inventories required to support production activities, 1['Materials Requirements Planning (MRP) is a methodology for deter- mining quantities and timing of up-coming deliveries of materials and components. MRP combines ordered and on-hand quantities with planned production requirements, taking into account production lot sizes and delivery lead times [Orlicky (1975); Wallace and Daugherty (1987)]. 54 status information is up-dated frequently (daily or weekly) in MRP systems. Manufacturing managers who use timely MRP inventory and pro- duction status data are likely to want their cost data to be current and coordinated with MRP data. DeWelt points out that traditional account- ing systems typically have a monthly time frame but MRP systems require that information be up-dated daily or weekly, because the consequences of being unaware of resource status can be very costly. In such cir- cumstances, failure to provide up-to-date cost data may diminish the ability of managers to interpret the cost consequences of their resource decisions. W This publication brought together N.A.A. Bulletins 11 through 15, plus 22, which were published originally during the 1950s. A section from pages 7 to 13, entitled ”Standard Costs for Cost Control," is sum- marized here. WW3- Operation control is acknowledged to be a precedent for cost control. By addressing appro- priate questions, standard costs can be applied to operation controls to help achieve cost control. Questions addressed by standard cost systems include: a Cost objective: for what are the costs incurred? m Cost accountability: who is responsible for control of the costs? Cost objectives and control objectives are related. Cost objec- tives are purposes for which costs are measured; control objectives are purposes for which costs are incurred. Manufacturing managers' actions 55 are guided by control objectives and are evaluated according to cost objectives. Cost accountability is a combination of cost objectives and control objectives. Cost control is a matter for individuals to address. Therefore, a standard cost system classifies costs according to the organizational structure of cost management responsibility, making individual cost control possible: Experience shows that control is most effective when standards are set in terms of personal responsibility for each cost incurred. Actual results are then measured against the respective standards in order that each individual may know how his performance compared with that which was expected [N.A.A. (1974, p. 10)]. Standard costs were important to the development of cost accounting procedures as a means of cost control. Engineers' specification of quantities of materials and labor hours could be incorporated into accounting measures through standard costs. This simplified the task of providing pre-determined costs where needed and assessing actual costs of production. Cost accounting concepts and measures became more for- malized through the influence of standard cost measures and variance re- porting procedures. Furthermore, standard costs systems helped to coordinate cost accounting data with engineering data, which is a cru- cial aspect of integrated manufacturing data systems. THE ROLE OF THE CONTROLLER Simon et al. (1954) reports the findings of a field study of the role and responsibilities of controllers in business organizations. Controllers, accountants and operating executives of seven companies WI on 0-“ 56 were interviewed to provide the data for the study. Interviews focused on (1) the organization and scope of the controllership function; (2) the structure, content, and distribution of accounting reports; and (3) patterns of communication within the controller's department and with operating departments [Simon et a1. (1954, p. vii)]. Three topics were especially relevant to the present study: accounting activities for reporting cost data, impediments to acceptance of cost data, and cost data measured in physical terms. W The Simon et al. (1954, p. 3) study classified controllers' duties into three activities: score-card, attention-directing and problem- solving. figg;§;g§;g activities measure and enter transactions. Accord- ing to the study, controllers, concerned about the accuracy of account- ing data, want valid transaction entry procedures to ensure the integ- rity of recorded data [Simon et al. (1954, p. 29)]. When data integrity is lacking, resource managers lose confidence in the data, become skep- tical about its implications and may make unwarranted, wasteful decis- ions [Simon et al. (1954, p. 32)]. A;§gn§12n;di;gg§1ng activities involve analysis of recorded cost data to discover evidence of operational problems or data errors. The Simon study found that cost data have more attention-directing value when the data convey information not easily obtained from other sources: Supervisors up to factory department heads use accounting reports for attention-directing purposes largely in areas that are not easily visible in the course of day-to-day supervision [Simon et a1. (1954, p. 26)]. 57 What can be interpreted from this finding is that accounting data do not necessarily draw attention more easily or more quickly than other func- tional sources, but accounting data can affirm what is reported by other sources. This has relevance to a point made later in the present study about reporting affirmative cost information. anhlgn;§glging activities are those where controllers offer their advice, participate in planning, or take corrective actions. Controllers interpret recorded cost data or provide prospective analytical cost data. As reported in the Simon study, "the operating executive has special needs for periodic accounting reports on items that are not visible from direct, day-to-day supervision" [Simon et al. (1954, p. 27)]. An interpretation of these Simon study findings is that accountants can interpret and report both the operational and accounting view of business operations and thus have a knowledge advantage. A key purpose of cost accounting activities is to transfer this knowledge advantage to resource managers who then might be more effective at managing costs. W The Simon study dealt with impediments to acceptance of accounting data by operating managers. The researchers found that managers accept accounting data if the managers believe the data are authentic. For example, when standard cost variances are reported to managers for explanation, they: 58 . were inclined to accept a standard to the extent that they were satisfied that the data were accurately recorded, that the standard level was reasonably attainable, and that the variables it measured were controllable by them [Simon et al. (1954, p. 29)]. On the other hand, the researchers found that if managers doubt the integrity of accounting data, they tend to ignore the data in assessing their own performance, unless they are forced to act on it to defend their position with superiors: When there were doubts as to the accuracy of recording or classification of data, when the factors causing variances were thought to be beyond their control, the executives simply did not believe that the standard validly measured their performance. Then they were influenced by it only to the extent that they were forced to think about the reactions of their superiors [Simon et a1. (1954, p. 29)]. Reassignment of cost data may cause managers to doubt the integrity of accounting data. The research team reported on managers' objections to cost assignment: Some objected to oversimplified determinants of standards that failed to account for important external factors causing variability in costs. A second major source of distrust of accounting standards was the recirculation of indirect costs, on two counts. First, operating people dislike having their statements include cost items regarded as not within their control, especially when such items trigger variances. Second, for recirculated indirect cost items admittedly par- tially controllable, the accuracy of the charges was doubted [Simon et al. (1954, p. 30).] These Simon study findings underscore the vital importance of data integrity to sustain managers' confidence on cost data. Managers are more confident about cost data they believe reflect the underlying reality of resource events that drive costs.15 15For another source on impediments to the use of information, see Mintzberg (1977). 59 WW: The Simon study found that receptivity to accounting data is en- hanced if users can interpret the data in physical terms: In those companies where the products can be measured at least roughly in physical units, manufacturing executives make more use of data expressed in physical units than data measured in dollars. Dollar comparisons are made largely in those situa- tions where there is no other common denominator for comparing production or inventory totals [Simon et a1. (1954, p. 31)] The Simon study findings suggest several key insights about resource managers' perceptions: m Managers are more likely to use cost data if they believe the cost data are authentically and accurately reflect underlying resource events. Managers may either ignore or be defensive about data perceived to lack integrity. a Managers may distrust recirculated cost data if they are suspi- cious that cost allocations obscure underlying resource events or cost drivers. I Accounting data that can be related to physical units of measure seems to be more authentic and acceptable because the data more clearly reflect operational realities. SUMMARY: TRADITIONAL COST ACCOUNTING LITERATURE Several points were gleaned from the literature on traditional cost accounting. s Cost accounting became a formalized means of identifying and classifying costs of production during the historic period of mass production development in the early part of this century. Cost assignment focused on direct materials and labor because materials and workers were the primary resource controlled by production management. 60 m Overhead pools and allocation methods developed as production activities became more complex and additional indirect costs of supervision and accounting became significant. These pooled indirect production costs were assigned to products based on labor because they were part of the cost of managing labor. I Cost centers were established to improve control over special- ized production areas, and eventually cost centers were also established to control burden costs. a After a period of time, financial accountants began to appre- ciate the objectivity of cost measures for inventory valuation, and procedures for attaching costs to products became commonly accepted. a Standard cost systems helped formalize cost accounting concepts regarding direct versus indirect costs and helped to coordinate cost measures with engineering data systems. I Classifications of cost accounting activities from the Simon et a1. (1954) field study support the role of cost accounting as described in this study. Historic cost accounting literature explains how cost accounting activities developed in traditional manufacturing. Cost accounts were established and classified to help management monitor costs of complex production based on engineering data; the classifications were labor- oriented because the worker was a key resource to manage. Cost data became accepted as the primary basis for inventory determination. The Simon et a1. (1954) study revealed that a controller's participative role had developed through using cost data to provide score-keeping measures of production output, direct attention to cost control pro- blems, and provide analytical cost data to help solve problems. LITERATURE ON CIM AND ITS IMPACT ON COST ACCOUNTING This section summarizes: (1) literature relevant to the charac- teristics of CIM described in Chapter II and (2) recent accounting literature about the impact of CIM on the role of cost accounting. Both 61 of these groups of literature were reviewed for theoretical development of hypotheses and hypothesis focus points presented at the end of this chapter. LITERATURE ON CIM CHARACTERISTICS CIM literature reflects the recency of CIM in the manufacturing environment. Most of the CIM literature discusses concepts and implica- tions in broad terms. Literature on applied situations is not exten- sive. The CIM literature generally suggests, but does not confirm, im- plications of CIM implementation. Nevertheless, the literature does provide some useful insights about the likely impact of CIM on cost ac- counting activities. A variety of literature describes CIM characteristics. Some of the literature addresses CIM's technical aspects and operational advantages; this literature includes Goldhar and Jelinek (1983 and 1985), Groover and Zimmers (1984), Gunn (1982), and Stover (1984). Literature con- cerned with accounting implications of CIM includes Cooper and Kaplan (1988), Dilts and Russell (1985), Johnson (1987), Johnson and Kaplan (1987), Howell et a1. (1987), and McNair, Mosconi and Norris (1988). This literature plus others have been examined for additional theoreti- cal support for the hypotheses and focus areas of the study.16 16Other literature examined but not summarized in this section includes Bruns and Kaplan (1987), Bylinsky (1983), Cohen and Zysman (1987), Ettlie (1988), Hill (1985), Richardson (1988), and Skinner (1985). 62 W Based on Goldhar and Jelinek (1983 and 1985), Groover and Zimmers (1984), Gunn (1982), Harrington (1984), and Stover (1984) characteris- tics and strategic advantages of CIM are summarized as follows: CIM is controlled by the flow of information between flexible, product-focused process centers. This is in contrast with traditional manufacturing, which is organized by the flow of materials between dedicated process areas. CIM integrates information and computer-aided technology with materials characteristics and enables the formation of cells to focus on product attributes rather than process attributes. Improved information and flexible processes reduces the need for buffer inventories near process centers to cover for uncertain- ties and disruption. Elimination of unnecessary buffer inven- tories can reduce waste, scrap damage, inventory risks and inventory carrying costs. With CIM, competitive advantage is not confined to economies of scale but can also include economic advantages of variety, flexible response to changing market demands, delivery reliabil- ity, high output quality, as well as cost competitiveness. CIM technology allows sophisticated, inexpensive inspection as well as quick, low-cost process changeovers and set-ups. CIM plant layouts are not organized with large areas of special- ized processes (stamping, welding, grinding, etc.). Product- focused, cellular configurations of flexible equipment using work crews are cost effective. Support costs for design, maintenance, and status information are significant in CIM; many such support costs are incurred in advance of and separate from the production activities that benefit from them. “Wan: The preceding points summarize characteristics and advantages of CIM as described in CIM literature. The same literature was examined closely to interpret CIM's influence on cost accounting. This section 63 presents several interpretations developed from a synthesis of the CIM literature cited previously. Since CIM is organized into product-focused process areas, tradi- tional patterns of production flow are changed. This has a bearing on cost identification. Costs in CIM can be identified more effectively with revised account names and regrouped accounts to match resource classifications suitable for CIM. By regrouping accounts for both resources and resource support, cost assignments can be made directly to product groups. Furthermore, since CIM depends on timely production flow through integrated processes, the consequences of disruption and delay can be costly. Instances of disruption or delay may be recognized and captured as a routine part of cost identification and entry. Steps are taken in CIM to reduce the time required for process set- ups, so that production can be run in smaller lot sizes and still be virtually continuous. Variety in production affects cost assignment since more variety exists in manufacturing processes, which can compli- cate the assignment of costs. CIM generates more indirect costs than traditional manufacturing, some of which precede production by such an extent that it is difficult to attribute it to production. This also complicates the cost assignment task. It is not unusual in CIM occasionally to have idle equipment. In- tegrated processes require that process cells produce output only when needed (not in advance) and in balance with other process areas. Period-oriented depreciation and other accounting charges are, there- fore, less valid in CIM. Traditional measures of utilization may show 64 inefficiencies in sub-units while the plant operations as a whole may be cost effective. Information is used extensively in CIM, so extensively that it is actually a resource or at least a significant cost driver. Since CIM's information needs can be costly, separate measures of information costs may be needed for direct assignment of information costs to resource groups. RECENT COST ACCOUNTING LITERATURE Recent literature on the impact of CIM on cost accounting has ques- tioned the relevancy of costs determined by traditional cost accounting systems. Some literature has focused on a lack of appropriate cost measures. For example, Kaplan (1983) says cost accounting's implicit model of manufacturing operations fails to recognize key changes in manufacturing management objectives brought about by CIM. Kaplan says innovative cost measures are needed if cost management systems are to provide feedback responsive to CIM. In similar fashion, other contem- porary literature, including CAM-I (1985 and 1986), Cooper and Kaplan (1988), Dilts and Russell (1985), Johnson (1987), Johnson and Kaplan (1987), and Miller and Vollmann (1985) criticize the relevancy of cost accounting practices in CIM, implying they are no longer supportive of management objectives. Still other literature, including Bruns and Kaplan (1987), Howell et al. (1987), McNair, Mosconi and Norris (1988), and Richardson (1988) focus on implementation of improved cost account- ing practices, based on research carried out at individual companies. 65 Out of this extensive body of recent cost accounting literature, CAM-I (1986, 1985), Cooper and Kaplan (1988), Dilts and Russell (1985), Johnson and Kaplan (1987), Johnson (1987), Kaplan (1983, 1985a) and Miller and Vollmann (1985) are representative sources.17 WM). In 1985, Computer-Aided Manufacturing International (CAM-I), an industrial research organization, initiated a cost management systems (CMS) development project to “identify a body of knowledge and to in- fluence practices in environments with advanced manufacturing technol- ogy" [CAM-I (1985)]. Two CAM-I documents summarize the aims of the CMS prOJectt WW (1985) and W Syg5gn_(§fl§l_£;gjgg§_£;g§ngggng (1986). These documents include several points useful for theory about cost management in CIM; the points are summarized in the following paragraphs. AIghi;gg§nrnl_§n9;§ggm1ng§. CAM-I observed that in traditional manufacturing (1) a relatively small number of parts or products were produced, (2) lot sizes were large in order to be economical, and (3) manufacturing processes were labor intensive. Since CIM differs from traditional manufacturing in all three respects, CAM-I (1985, p. 3) con- cluded that traditional cost accounting concepts and practices have "architectural" shortcomings. According to CAM-I, today's cost manage- ment systems continue to have cost tracing objectives of traditional 17Additional relevant literature on contemporary cost accounting practices include Eiler (1986), Hall (1983), Johannson (1984 and 1985), Porter (1985), Seed (1984), and Wingard (1985). This literature is consistent with the literature reviewed in this section. 66 manufacturing and lack a good fit with cost drivers and control aspects of CIM. In addition, CAM-I says the role of cost accounting in tradi- tional manufacturing is one of passive historic review, while CIM re- quires an active cost management system, integrated with manufacturing process control. WW- Accordins to CAM-I (1985). traditional cost accounting systems do not measure demands on production resources precisely enough. CIM processes handle a variety of product designs and parts, and greater variety of product designs and parts means less consistency in demands on production resources. CAM-I notes that some mature products may have reached the stage of their life cycle where demand on resources is consistent and reduced to a minimum; others may be in the introductory phase of their life cycle where demand on resources is inconsistent and not yet minimized. In addition, complex- ity of product features can vary demand on production resources. Some products may be marketed with attractive, but costly, options; others may be marketed with straight forward, simple designs. For all these reasons, it is more important now that cost management systems have measures of differing levels of demands on resources [CAM-I (1986)]. CAM-1's comments underscore a theoretical point argued in this study: identify costs in relation to the resource events that cause the consumption or use of resources. Failure to measure consumption or use of manufacturing resources with reasonable precision can seriously jeopardize management's ability to assess their competitive position. Managers who perceive the costs of such products to be more than jus- tified by demands on manufacturing resources may fail to compete 67 aggressively with high volume, high quality products. Furthermore, managers may introduce new products without a full awareness of how costly the products will to be in terms of demands on manufacturing resources . 18 Rg§n9§g_§g§;nn_11ng§. Traditional cost accounting practices were developed in manufacturing environments where production lot-sizes and intervals between set-ups were planned to minimize process interrup- tions. Set-ups and tear-downs were time consuming and costly to per- form. CIM technology has reduced set-up times and costs substantially. Set-up times are reduced by changing fixtures and improving fasteners to speed up the set-up tasks, or by dedicating machines to processes there- by eliminating set-ups. On computer-aided equipment, set-ups can be installed with great precision and, therefore, require relatively few "try-out” parts. With reduced set-up times, small production lot sizes can be produced economically. Work-in-process inventory is reduced sub- stantially, freeing up floor space for more capacity. Set-ups and tear- downs occur more frequently, but require less time with each occurrence. In traditional manufacturing, set-up times are typically not speci- fied as direct labor on operations routing sheets. Therefore cost accounting systems generally do not measure costs of set-ups directly, but record set-up costs in manufacturing overhead. However, managing set-ups gains importance in CIM, which suggests that more specific cost accounting measures of set-ups may be warranted. 18See Johnson and Kaplan (1987) and Worthy (1987) for further discussion of this point. 68 Lgag_Dgngnggngg_2n_nirggt_Lahgr. Much of traditional manufacturing is relatively labor-intensive, either in terms of labor content in product costs, or in terms of dependence of production processes on workers' skills. CIM is technology-intensive more than labor-intensive. In a CIM environment, direct labor costs are reduced while equipment, technology and support costs are increased. fligh_QggIhg§g_En§§§. CIM costs include computer-aided equipment and tooling, computer-aided testing, computer-controlled materials han- dling, and information from computer data base systems. A variety of other technology-related support costs are incurred, including costs of software and salary costs of technical and planning employees. All of these technology and support costs are typically recorded in manufac- turing overhead accounts. Therefore, CIM plants have high overhead ap- plication rates. According to CAM-I (1986), overhead costs take on the appearance of being unmanageable, and managers may be motivated to take unwise actions to avoid overhead charges. In fact, ”overhead" as a single classification of indirect production costs is less appropriate in CIM. Multiple overhead pools may be more appropriate. Qg§§_pnixngflgln;19n§hin§. Traditional cost accounting concepts are built on two specific cost categories, direct materials and direct labor, and a generalized cost category, manufacturing overhead. CIM's advanced technologies involve additional significant cost drivers which do not relate strongly with direct materials, direct labor, or with production volume. Direct labor has traditionally been measured very closely, in part due to a high number of job classifications and the existence of labor incentive pay schemes. In CIM, fewer physical tasks 69 are performed by workers. Consequently, fewer job classifications usually are found in CIM than in traditional manufacturing. Workers' tasks are more versatile. Workers guide or monitor manufacturing pro- cesses, utilizing computer-aided information, tooling, and materials handling equipment. Furthermore, labor is typically performed in small groups at various manufacturing cells and, therefore, it is difficult to associate production output with individual workers.19 Incentive pay schemes are appropriate where workers can exercise individual control over the pace of production or the volume of output. However, workers tending automated processes have little control over the pace of production or volume of output; efforts of individuals cannot be readily associated with production output. Wages of such individuals may be classified as indirect labor. Advanced technology, therefore, is likely to reduce the number of jobs that can be classified as direct labor jobs. For this reason, assignment of overhead costs based on direct labor is unlikely to represent demands on production resources. I;ngk1ng_£n1§§. CIM's integrated cellular arrangements of work areas and reduced work-in-process inventories actually reduces the need for tracking production status with entering detailed data. Fewer costs are traced through the flow of production. Where data entry is re- quired, in CIM parts can be identified with magnetized bar code strips, 19Where automation is introduced the number of jobs that can be measured as direct labor jobs is reduced. Carlson (1982) reports declining use of incentive pay schemes for individuals' wages in in- dustries where automation is introduced into production processes. 70 which can be read easily and accurately for entry into production con- trol data systems. The ”architectural“ shortcomings noted in CAM-I (1985, p. 3) sup- port the cost identification hypothesis and focus points later in the chapter. Without changes in account classifications and groups and in the designated boundaries of cost centers, cost systems fail to take into account the variety of resource events, the different character of set-ups, the diminished dependence on direct labor and correspondingly high overhead rates, and the existence of new cost driver relationships. CAM-I also observed that there are opportunities in CIM to improve classifications of overhead costs and improve the tracking of materials and components through production. W This research is based on visits to more than 20 firms, where the authors found that traditional cost measures distort product costs by understating costs of low volume products and overstating costs of high volume products. The reason is the use of volume-based cost alloca- tions, which fail to recognize the differential costs of transactions per unit of product. The article recommends the following: (1) Recognize variable costs at transaction points, since many variable costs are a function of the number of transactions the products require, and (2) Allocate costs from cost pools by using multiple allocation bases that suitably reflect cost drivers rather than using vol- ume-based allocation factors. 71 These recommendations support two themes emphasized in this study: that transaction triggers be chosen to measures costs based on consump- tion or use of resources and that individual cost pools be used for major classes of resources. W In explaining the flexible processes of CIM, Dilts and Russell (1985) describes several manufacturing characteristics that are relevant to cost accounting systems, paraphrased below: Manufacturing processes are accurate and reliable and provide consistently high quality of output. Computer-aided machines or processes are easy to set up and require fewer ”set-up tryout" parts. Long production runs are no longer required to minimize overall costs because set-up times and costs are reduced. Because computer-aided machines are versatile, machine utiliza- tion can be increased. Furthermore, more output variety can be achieved without reducing capacity utilization.2 Since less direct labor is required for manufacturing processes, process throughput time is reduced. Consequently, work-in- process inventories and overall space requirements are both reduced. Growth can be attained without expensive capacity expansion. All of the preceding attributes combined suggests that there is a wide range of manufacturing volumes at which CIM's process areas are economically cost effective. CIM production focuses on variety and flexibility, whereas traditional manufacturing strives for stabilized production. The attributes of flexible manufacturing described in Dilts and Russell (1985) are strikingly different than the stabilized production 20Recent research casts some doubt that machine utilization is increased by CIM technology [Jaikumar (1986); Hayes and Jaikumar (1988)]. 72 characteristics upon which traditional costing concepts are based. The difference in attributes underlies a major theoretical contention of the study: that cost measures and assignment should focus on consumption or usage of direct resources. I I! v I As the title suggests, the aim of Johnson and Kaplan's Rglgxnngg Lag; is to explain why cost accounting systems fail to help managers identify costs of their processes and products. Johnson and Kaplan say collection and reporting of cost data are driven by financial accounting reporting and disclosure requirements and as a result "accounting infor- mation is too late, too aggregated and too distorted for use by managers in making their planning and control decisions” [Johnson and Kaplan (1987, p. 1)]. Johnson and Kaplan say the loss of relevance has three consequences: (1) Accounting reports currently provided are not sufficiently helpful for cost management purposes. The reports do not aid efforts to improve productivity or reduce costs, and in fact may actually distract attention from critical cost management factors: By not providing timely and detailed information on process efficiencies or by focusing on [relatively insignificant] inputs . . . such as direct labor . . . the management ac- counting system not only fails to provide relevant information to managers, but it also distracts their attention from fac- tors that are critical for production efficiencies [Johnson and Kaplan (1987, p. 2)]. (2) Product costs provided to managers for their decisions are not accurate, primarily because of the use of simple, arbitrary measures 73 which do not accurately measure demands made by products on the firm's resources. (3) Because of strong influence brought on accounting practices by financial reporting requirements, the planning and control horizon of managers contracts to a short-term monthly cycle of financial account- ing. A more appropriate horizon for decision making is the life cycle of products and processes. In the long run, businesses must manage effectively throughout the entire life cycle of products and manufac- turing processes. These points support the cost identification, entry and reporting hypotheses that cost accounts and transaction measures in CIM will be revised to identify and report costs of resource events more explicitly and with a shorter time horizon than has been customary in traditional manufacturing. One serious shortcoming cited by Johnson and Kaplan is "cross-sub- sidies" among products, where some products appear to be more costly and others less costly because demands on production resources are not well identified: Although simplistic product costing methods are adequate for financial reporting requirements -- [to] yield values for inventory, cost of goods sold to satisfy external reporting and auditing requirements -- the methods systematically bias and distort costs of individual products. The standard pro- duct cost systems typical of most organizations usually lead to enormous cross subsidies across products. When such dis- torted information represents the only available data on "product costs,” the danger exists for misguided decisions on product pricing, product sourcing, product mix, and responses to rival products [Johnson and Kaplan (1987, p.2)]. This observation supports the cost assignment hypothesis that cost pools will be sub-divided and assigned according to resource group. 74 Johnson and Kaplan say cost systems can collect more exact cost data by taking advantage of advanced information technology for data collection and recording. Data too difficult or costly to collect in a traditional environment can now be collected more easily and relatively inexpensively. In fact, environments in which advanced technology is employed in manufacturing planning and process control offer the oppor- tunity for greater use of real-time data. Simplified, aggregate procedures, adopted in earlier decades because more relevant and timely procedures would have been too costly or even infeasible, no longer need to be tolerated. The computing revolution of the past two decades has so re- duced information collection and processing costs that vir- tually all technical barriers to the design and implementation of effective management accounting systems have been removed. The increased complexity of operations in today's global, technological economy has been matched by a corresponding increase in the capabilities of systems to provide relevant and timely information on the operations [Johnson and Kaplan (1987, pp.5-6)]. The cost reporting hypothesis is based on the assumption that cost management systems of CIM firms will use advanced technology to collect and maintain cost data. Resource managers may actually prefer on-line inquiry to paper documentation since on-line inquiry provides a "real time" grasp of costs of resource sacrifices at points of occurrence. W Johnson distinguishes between the terms ”cost accounting” and ”cost management." Cost ngggnnging concentrates on attaching costs to pro- ducts for inventory valuation. Cost nannggngnt focuses on using resour- ces cost effectively to accomplish production objectives. Since these are not equivalent management objectives, cost accounting and cost management developed along separate paths. But, Johnson says, since 75 about thirty years ago cost nggnnnging information has been used for cost nannggngn; purposes. Johnson questions the wisdom of assuming that cost accounting concepts and practices can satisfy cost management objectives. One source of difference between cost accounting and cost manage- ment objectives is the accounting treatment of indirect manufacturing costs. When manufacturing output is diverse, costing procedures for attaching indirect costs to units of product can produce misleading results. With diverse output, demands are placed on manufacturing resources at varying rates, which may not be reflected in overhead ab- sorption rates based on direct labor. Cost accounting treats indirect costs in the least costly manner possible--aggregating them in a few cost pools and usually attaching them to products with a single denominator. Cost management, on the other hand, requires that indirect costs be traced carefully to the consumption of specific resources that cause costs. Product cost accounting informa- tion, therefore, is not useful for product cost management if there is any diversity among products [Johnson (1987, p. 8)]. This comment from Johnson (1987) supports the cost assignment hypothesis. Unless indirect production costs are pooled separately by class of resource, such as machinery, tooling, or technology, overhead pools would contain a mixture of costs of resources (machinery, tooling, technology) and costs of resource support. 9 9 4 9 4 Like other literature on CIM and cost accounting, Kaplan's articles discuss traditional costing assumptions, now outmoded, about the nature of manufacturing processes and activities in CIM. He argues for changed measures of manufacturing cost performance, including new measures for 76 quality, flexibility, inventory performance, innovation and productivi- ty. Reasons for inadequate cost measures are suggested by the theory developed in this study. If inadequate measures exist in cost manage- ment systems, ambiguous motives for the measures may be the cause. Given that "cost" is defined as a measure of productive (non-wasteful) resource sacrifice, costs are identified based on events where resource sacrifices are observable and where accountability of resource managers can be established. There are fairly specific measures for financial reporting requirements and for "audit trail" substantiation, for which the motives are relatively unambiguous. But cost management motives have been articulated less clearly in the past and, consequently, meas- ures for effective cost management are not well developed. One response to Kaplan's call for improved measures is to shift the focus of produc- tion cost identification from cost attaching attributes to resource sacrificing attributes. M Kaplan (1985a) describes visits made to four leading manufacturing firms where innovative manufacturing technology had been implemented. He expected to find their cost accounting practices to be modified to correspond with manufacturing innovations. Instead, he found a lag: innovations in manufacturing had occurred but the expected corresponding cost accounting adaptations had not. Cost accounting practices, firmly established to provide information for financial reporting, were slow to adapt to information requirements of integrated manufacturing. 77 One company's cost accounting systems aggregated its accounting numbers in ways not comparable to its reorganized manufacturing opera- tions. Another company's traditional standard cost system focused on labor costs even though labor had become an insignificant cost element; that company was forced to calculate its actual product costs informal- ly, outside the accounting system. Another company's indirect costs were a large proportion (and its direct labor costs a low proportion) of its total production costs. Its extremely high burden rate distorted its cost allocations and make versus buy decisions. A fourth company continued to employ outmoded accounting performance measures that did not reflect major changes in product characteristics, process technology or the market environment. Kaplan's findings indicate that companies with advanced manufacturing technologies and integrated manufacturing processes may be slow to adapt their cost accounting measures and prac- tices to correspond with the data features of CIM. In interpreting Kaplan (1985a), only recently has cost management become a crucial issue to manufacturers. This may explain why Kaplan found adaptation of cost accounting systems lagging behind manufacturing modernization efforts. While the concern appears to be one of obsoles- cence in cost accounting practices, the response is likely to emerge from re-directing the focus of cost management systems from tracking costs toward just what the name implies: managing costs, where costs are identified based on appropriate resource events. When costs are well identified, wasteful events can be identified and eliminated, and cost accountability can be maintained effectively. 78 W Even though control of manufacturing overhead costs is a leading cost management issue, cost accounting models allocate overhead costs but do not explain them, according to Miller and Vollmann (1985). This is because most of the driving force behind overhead costs is not in direct labor nor units of production volume. Likewise, overhead costs may correlate with units of volume, but volume is not what is driving overhead costs. The bulk of overhead costs are derived in a ”hidden factory" of transaction sources [Miller and Vollmann (1985, pp. 144, 146)]: a logical transactions which execute or confirm movement of materials; a balancing transactions to ensure that supplies of materials, labor and technological capacity are equal to demand; a quality transactions which identify and communicate specifica- tions, certifications, etc.; and I change transactions which update manufacturing information systems for changes in engineering designs, materials specifica- tions, bills of material, standards, routings, and schedules. The article goes on to say that forces that drive these various transactions need to be identified for cost management purposes. By careful analysis, it can be possible to eliminate some transactions without restricting production. Stabilizing the manufacturing environ- ment reduces transactions and errors, and also eliminates transactions required to correct discovered errors. 79 WW Contemporary cost accounting literature generally supports Chapter II's theoretical analysis of changes in the role of cost accounting, as indicated in the following summary: CAM-I (1985 and 1986) identified "architectural” shortcomings in traditional cost accounting's model of manufacturing, including failure to take into account the variety that now exists in CIM, the different character of set-ups, a diminished dependence on direct labor, correspondingly high overhead rates, and the existence of new cost driver relationships. Dilts and Russell (1985) describe characteristics of CIM that are relevant to cost measures and objectives. Johnson and Kaplan (1987) explain cost accounting's loss of relevance by saying that the shortcomings in cost accounting's model result in data that is too aggregated to reflect demands on resources. Consequently, product costs are imprecisely measured, resulting in cross-subsidies among products. Johnson (1987) distinguishes between cost accounting and cost management. The former deals with determination of unit costs of products for financial reporting, and the latter deals with effective use of production resources to meet management objec- tives. Since costing for cost accounting has dominated over cost management, there are areas of inadequacy in cost account- ing concepts and practices, especially for indirect production costs. Kaplan's series of articles trace the history of developments in manufacturing that have led to today's inadequacies. His report on visits to four plants reveals how difficult it can be to bring about changes in cost accounting activities. COOper and Kaplan (1988) describes how product costs are dis- torted by the use of volume-based measures and allocations. They recommend transaction based measures and multiple alloca- tion bases. Both recommendations are similar to the ideas about choice of transaction triggers and resource oriented cost al- locations described in this study. Miller and Vollmann (1985) describes how the role of information impacts overhead costs because of the extensive amount of sup- port that advanced technology resources require. The shortcomings described by CAM-I, Johnson, Kaplan and the others can be rectified by improved cost identification and entry based on 80 characteristics of manufacturing resource events in CIM, and by clear linkage between cost assignment bases and cost objectives, upon which cost reporting is based. Improvements like these are the basis for the research hypotheses. RESEARCH HYPOTHESES AND FOCUS POINTS Four research hypotheses identified originally in Chapter I are presented again in this section and linked with theory delineated from the literature. Major theoretical points are also developed into “focus points” to accompany the hypotheses. The focus points are used later in Chapter V to organize the study findings, support the hypotheses, and amplify significant aspects of the research hypotheses. There are fourteen focus points, each numbered according to the hypothesis to which it relates. The focus points are listed in Table 10 in Chapter V. COST IDENTIFICATION A cost system needs pre-determined means of identifying and clas- sifying costs in accounts. Cost identification includes creation of account names consistent with the nature of resource costs, grouping accounts by class of resource, designation of cost center boundaries and determination of the management level of responsibility for the costs. The cost identification hypothesis is: Hypothesis 1: Compared with the traditional setting, CIM will lead to a greater number of cost centers, which will be smaller in size. In addition, classification and groupings of ac- counts will be based on a wider set of resource groups than materials, labor and overhead. 81 Traditional batch manufacturing is organized with large cost cen- ters that specialize in machine processes (press department, machining department, etc.) [Reeves and Turner (1972)]. CIM organizes clusters of flexible process areas that focus on attributes of products rather than machine processes [Goldhar and Jelinek (1983 and 1985)]. Restructured manufacturing activities in CIM justify redefinition of cost centers and pools of cost accounts so that boundaries of cost responsibilities match patterns of consumption or use of resources; that is the justification for the cost identification hypothesis. In addition, several issues raised in the review of the literature pertain to cost identification and are detailed in the following focus points. 1.1 REDESIGNATION OF In CIM as contrasted with traditional manufactur- COST CENTERS ing, resources are regrouped to focus on product families and production flow patterns are restructured. Therefore, to retain the homogeneous character of cost center activities it is necessary to designate more cost centers, which will be smaller in size or area. It has been a long-standing concept in cost accounting that a cost center defines the boundaries of manager control over reasonably homo- geneous resource activities [Litterer (1961a)]. However, the strategic manufacturing objectives of CIM (flexibility, quality by design, deliv- ery reliability and cost management) change the organizational con- figuration of resources (the plant floor layout), restructures manufac- turing activities, and revises production flow patterns. Therefore, to continue to attain homogeneity with cost centers, more cost centers are established, though they are smaller in size (floor space occupied). 82 1.2 CLASSIFICATION OF In CIM as contrasted with traditional RESOURCES AND SUPPORT manufacturing, a substantial proportion of production costs relate to resources other than materials or workers; therefore additional classes of resour- ces and support are carved out of traditional overhead cost pools. For historic reasons, cost identification in traditional manufac- turing focused on materials and labor [Chandler (1977), Henrici (1960)]. In CIM, many technological costs are not strongly related to materials or labor [Miller and Vollmann (1985)]. In traditional cost systems, costs of non-labor resources, such as machinery and equipment, tooling, and technology and information, are classified in manufacturing overhead pools but in CIM these costs will be more directly recognized as resource costs. In addition, other indirect costs will be classified into resource support pools. 1.3 IDENTIFICATION OF COST In CIM as contrasted with traditional MANAGEMENT RESPONSIBILITY manufacturing, some costs are incurred by individual cost center but are managed at a higher integrative responsibility level; thus, such costs although traceable to individual cost centers are assigned to the higher responsibility level cost centers. One aspect of integrated manufacturing processes is that efficiency at the local (micro) level of work stations or cells does not necessari- ly imply overall efficiency at a higher macro level, such as for a product group or an entire plant [Goldhar and Jelinek (1985)]. Cost of local work stations or cells in CIM are likely to be monitored within a higher macro-level context [Johnson (1987)]. It is conceivable, there- fore, that a cost may be traceable to a local work station or cell but in fact be the responsibility of a manager at a higher integrated level. 83 If so, the cost will be recognized at the local level and identified with the higher level cost center. 1.4 CLASSIFICATION OF In CIM as contrasted with traditional manu- DIRECT COSTS facturing, some worker-related and technology- related costs traditionally classified as indirect are now classified as direct; examples include costs of set-ups, inspection and maintenance. Managers in the field prefer direct costs over redistributed costs [Simon et al. (1954)]. Given the previous point about the establishing of additional classes of direct resources, in CIM more traditionally in- direct costs will be classified as direct costs. An example is the labor cost of performing machine set-ups, which likely will be clas- sified as direct labor costs in CIM. COST ENTRY The second hypothesis concerns the choice of "trigger" resource events that can be used to capture cost data for cost entry. The flexi- ble nature of CIM activities and the changes hypothesized for cost responsibility boundaries suggest that a different set of cost transac- tion triggers will be used in CIM. Therefore, the hypothesis is: Hypothesis 2: The events critical to effective CIM production management are different than the events that "trigger" cost account- ing data in traditional costing systems. Therefore, cost transaction triggers in CIM will be greater in number and diversity than those of the traditional setting. The aim in selecting resource events as transaction triggers is to capture costs where the relevant resource activities occur. This is a more important issue in CIM because of the flexibility and variety that 84 exists in pursuing the strategic manufacturing objectives. Several issues regarding cost transaction triggers are developed into four focus points. 2.1 PLACING TRIGGERS CLOSE In CIM as contrasted with traditional manu- TO RESOURCE EVENTS facturing, strategic manufacturing objec- tives and CIM technology require that costs be captured more closely to the points where demands on resources can be identified, rather than at traditional points of entry/exit transfers or labor operations. Given that "cost" has been defined as a measure of resources con- sumed or used (gradual consumption over time), an ideal trigger event is one where resource consumption or usage can be captured. Moreover, the strategic manufacturing objectives of CIM suggests that there are numer- ous resource events available as triggers. Traditional manufacturing involves less close scrutiny of detailed resource events, and therefore costs generally are captured at transfer exit points. However, exit points may be too far removed from resources to be satisfactory as triggers in CIM [CAM-I (1985 and 1986)]. 2.2 RESPONSIBILITY FOR COST DATA In traditional plants, collecting -- A SENSE OF DATA OWNERSHIP and maintaining cost data is typi- cally an accounting responsibility, separate from responsibility for collecting and maintaining en- gineering or production data. In CIM plants, collection of both production and cost data is largely merged into one data collection system with the result that resource managers have a greater sense of ”ownership" of cost measurements. Historically, cost accounting evolved to provide monitoring capa- bility to manufacturing, and cost data were considered part of manufac- turing data [Chandler (1977)]. Since then, other uses of cost data have made cost accounting a separate data collection system. Managers have 85 come to view cost data as “owned" by accounting, provided to them from accounting [Johnson (1987)]. However, CIM will bring about a reintegra- tion of data collection activities. This will renew in manufacturing managers a sense of ownership responsibility for collection and use of cost data. 2.3 TRIGGERS FOR EXPENDABLE In CIM as contrasted with traditional AND DURABLE RESOURCES manufacturing, triggers can distinguish between expendable resource consumption and durable resource usage. Costs of durable resources are as- signed on the basis of usage measures rather than period-oriented charges for capacity provided. Durable resources generally have been the source of production capacity in manufacturing. A key objective has been to keep capacity utilization at a high level so that all fixed overhead capacity costs are absorbed into products [Johnson (1987)]. Accordingly, costs of machinery and equipment, tooling and other capital assets have been assigned as fixed, period-oriented capacity charges. However, the integrated flexibility of production in CIM, suggests that the basis for cost assignment of durable resources be changed from period-oriented capacity to nsnge. In order to accomplish this, events involving usage of durable resources will trigger cost transactions. 2.4 KEEPING COST In CIM as contrasted with traditional manufacturing, DATA CURRENT cost data will be kept more current by collecting data at trigger events where demands on resources occur, files will be up-dated more frequently, and a shorter plann- ing and control time horizon will apply. This focus point pertains to the relative currency of cost data, which depends on how frequently costs are captured. To meet its flexibility and delivery reliability objectives, CIM operates with a 86 shorter planning and control time horizon than the monthly time frame typical of traditional manufacturing [Goldhar and Jelinek (1983)]. Accordingly, CIM's triggers are expected to capture costs more frequent- ly than traditional cost triggers. COST ASSIGNMENT The third hypothesis concerns the linkage between identified costs and cost objectives for cost reporting. The hypothesized changes in cost identification (involving resource classifications) and transaction triggers lead to changes in the bases of cost assignment. In addition, there is a greater need in CIM to assign indirect support costs consis- tently with related direct resources, since support costs are substan- tial in CIM [Miller and Vollmann (1985)]. Accordingly, the third hypo- thesis is: Hypothesis 3: Cost assignment in CIM will be changed by explicitly relating support costs with resource costs and assigning both types of cost to cost objectives by class of resource, using relevant cost assignment factors. This is contrasted with traditional manufacturing where all over- head costs are typically assigned based on direct labor or some other single factor. A strong motive in CIM for reclassifying costs traditionally clas- sified as manufacturing overhead into classes of resources is that managers are dealing with more cost drivers, which are obscured if left in overhead. Also, many of the cost drivers do not relate strongly to direct labor and need to be assigned more directly or based on a more appropriate resource factor. Three focus points elaborate on these issues. 87 3.1 ASSIGNING SUPPORT COSTS WITH In CIM as contrasted with traditional RELEVANT RESOURCE FACTORS manufacturing, resource support costs are a more significant cost component and are assigned based on usage of relevant resources supported rather than based on traditional assignment bases (direct labor, machine hours, or direct materials). Costs are assigned to reflect the underlying relationship between resources consumed or used and production output [Porter (1985, ch. 3)]. Given the level of resource support in CIM, it is logical that accounts for support costs be grouped by class of resource (as suggested in Hypothesis 1). Assigning these costs will then be on the basis of a factor relevant to the class of resource supported. Aggregate costs can then be penetrated back to their source, or penetrate product costs to their underlying components, as discussed in focus point 3.3 below. 3.2 ASSIGNING COSTS BASED ON In CIM as contrasted with traditional COST MANAGEMENT OBJECTIVES manufacturing, cost drivers are more diverse. Also, costs are assigned with assignment factors that link cost objectives with a greater number of cost pools or classes rather than with traditional broad cost pools and a single factor for cost assignment. This focus point aims at the need to clarify the assignment of resource costs to cost objectives by linking resource activities more directly to cost objectives. Different cost objectives will exist in CIM to support strategic manufacturing objectives, and different cost assignment will be needed [Miller and Vollmann (1985)]. This focus point accompanies point 3.1 above, which relates resource support to resource classes; this focus point relates combined resource costs (direct resource costs and support) to cost objectives. 88 3.3 PENETRABLE COSTS WITH In CIM as contrasted with traditional TRANSPARENT COST ASSIGNMENT manufacturing, strategic manufacturing objectives present a greater need for penetrable cost data, which require transparent cost assignment procedures. Managers often want to trace costs to underlying cost drivers and, therefore, expect to be able to penetrate cost accounting data through cost assignment and classification schema. The Simon et a1. (1954) study found that cost data are more likely to be used by resource mana- gers if they perceive that the data clearly reflect the resource events for which they are responsible [Simon et al. (1954)]. ”Transparent“ is defined here to mean that the means of assigning costs do not obscure the source of the cost, and the user of cost data can "see through" the assignment to underlying cost drivers or resource events where the costs were incurred. “Penetrable” costs means assigned costs can be decom- posed backward to their underlying components. Penetrable is a cost attribute that depends on transparent cost assignment procedures.21 Given the variety of cost drivers existing in CIM, data transparency and cost penetration capabilities are likely to be more important to managers than in traditional manufacturing. COST REPORTING The fourth hypothesis concerns changes expected in reporting of costs in CIM: 21The source for these notions is an interview with a systems designer at one of the CIM sites. 89 Hypothesis 4: In the CIM setting, accounting information will be up- dated more frequently, “real-time“ query form of reporting will occur more frequently, and the focus of periodic accounting reports will be on strategic manufacturing objectives versus the financial product costing objectives of the traditional setting. Cost data that support CIM's strategic manufacturing objectives are current, readily available and focused toward cost objectives consistent with the manufacturing objectives. Focus points deal with cost report- ing issues involving supportive cost objectives, currency of data, and focusing on cost objectives at a macro level. 4.1 COST OBJECTIVES AND STRATEGIC In CIM as contrasted with traditional MANUFACTURING OBJECTIVES manufacturing, cost objectives will reflect a greater emphasis on strategic manufacturing objectives than on financial product cost- ing objectives. Traditional cost accounting has evolved to have a relatively strong orientation to product costing for financial reporting. CIM's cost management requirements are likely to re-orient cost objectives with more emphasis on forward looking planning and control, rather than summarizing past costs for financial reporting [Johnson and Kaplan (1987)]. 4.2 COST MANAGEMENT REPORTING In CIM as contrasted with traditional FOR COST MONITORING manufacturing, production monitoring will focus more on integrated costs at a macro level than is the case in traditional manufacturing, where cost monitoring focuses on the individual work center. Given the integrated nature of CIM's manufacturing activities, cost monitoring will necessarily be done at more of a macro level. Cost control will focus on local cost center levels, but cost efficiency and 90 capacity utilization will be assessed at a macro, integrated level [Cohen and Zysman (1987, pp. 165-67); Richardson (1988, pp. 90-91)]. The means of cost reporting and reporting formats will reflect these changes. 4.3 REPORTING COST CONSEQUENCES In CIM as contrasted with traditional OF DISRUPTION AND WASTE manufacturing, reported cost informa- tion is used to a greater extent to identify sources of production waste, either through variance analysis or through specific account recognition. Detection and elimination of waste is a point of emphasis in CIM, not only because of the direct cost of waste itself, but also because of the serious disruptive consequence of sources of waste [Harrington (1984); Goldhar and Jelinek (1985)].22 Reported costs will be used to a greater extent in CIM to help discover and eliminate the sources of V88 '28 . SUMMARY This concludes the specification of the research hypotheses and associated focus points for research analysis. Taken together, they can be summarized in the following list: I Costs are identified where resources are used or consumed; resource event patterns are different in CIM than in traditional manufacturing, leading to revised cost classifications, account groupings and cost center boundaries. s CIM cost transaction triggers will be based on events that reflect demands on resources, affirmation of resource objec- tives, or conformance to specifications. 22Waste is cost incurred that neither enhances nor maintains value. Waste can result from value-lost events or activities, such as scrap due to damage to inventory in transit or in storage. 91 a Cost assignment to products or cost centers in CIM is based on two classes of cost drivers: resources and resource support. Resource support will be classified in CIM by class of resource and assigned on the basis of measures of resources supported. a Cost reporting in CIM has cost management objectives as well as financial reporting objectives. For cost management purposes, cost data must be kept up-to-date, have a current time frame orientation, and be available for on-line inquiry. A field study design is specified in the next chapter to inves- tigate the four hypotheses and related focus points for the purpose of assessing the impact of CIM on the role of cost accounting. CHAPTER IV RESEARCH DESIGN INTRODUCTION Previous chapters have described the changes occurring in manufac- turing with regard to integration of manufacturing processes and the application of computer-aided technology. As stated in Chapter I, the basic research question is: How does the integration of manufacturing processes and the application of computer-aided technology affect the role of cost accounting in cost management systems? A field study design was determined to be appropriate for this question. The design included visits to both traditional and CIM plants where interviews were conducted with accounting, engineering and production personnel. This chapter specifies the field study design and describes the sites visited. In addition, the chapter describes the methodology used to gather and analyze the empirical data. The findings that resulted from these procedures are presented in Chapter V. The chapter begins with a discussion of research design objectives, followed by a section on site visits, and concludes with a description of data analysis procedures. RESEARCH DESIGN OBJECTIVES This study focuses on an emerging issue, which is still in an exploratory stage. Manufacturers are in the process of integrating their production activities and are experiencing significant organiza- tional changes in the process. Relevant theoretical constructs about the role of cost accounting in CIM are not well developed nor grounded 92 93 in knowledge about, or experience with, relationships among variables. This research can make a contribution by discovering in a field context descriptive findings useful as grounded theory for future research. Exploratory field research can discover patterns of descriptive find- ings, which may be articulated into testable propositions for further research [Kaplan (1985b); Yin (1981)]. Such exploratory research builds theoretical understanding, focusing on explanation as much as verifica- tion. Stone (1978) describes exploratory research as an early stage of a complete cycle of theory building and testing. As described in Stone (1978), the cycle begins with observation of real world phenomena from which explanations of apparent phenomena are induced. From the explana- tions, predictive hypotheses are deduced, which then are tested by rigorous comparison with real world phenomena. The results of the tests may return the researcher to the initial exploratory stage again, top of the circle again, with more phenomena to be explained, and so on. The evolving impact of CIM on the role of cost accounting is a research question that corresponds to the early exploratory stage of Stone's research cycle. Exploratory research is interested in processes reflected in the field context being investigated. Research on processes differs in certain respects from research involving measures of variance among variables [Mohr (1982)]. Process research examines interactive patterns of events driven by objectives, or "syndromes."23 Theory may suggest 23Seashore (1961) defines "syndromes" as identifiable patterns that reveal underlying generalizable characteristics or concurrent relation- ships among variables. 94 what causes the events to interact. Rather than proving cause, the main purpose of process research is to learn how the events interact and how that interaction is manifested [Mohr (1982)].24 THE DESIGN CHOICE -- A FIELD STUDY Given that the aims of this research are deductive exploration and explanation rather than proof or verification, a field study design is the appropriate design. Field studies can provide theoretical explana- tions from which verifiable hypotheses can be induced. Since the study findings are based on explanations obtained directly from a ”real world" context, theory deduced from the findings stand a good chance of induc- ing realistic testable hypotheses. In that sense, this research responds Hopwood's call for research on how accounting actually operates and in the context in which it is carried on [Hopwood (1983)]. Field studies are designed to capture contextual understanding. The design aims at discovery in two dimensions: a single incident dimen- sion revealed by anecdotal insight, and a more general dimension revealed in discovered patterns or trends [Yin (1984)]. The primary source of empirical data for this study is on-site interviews with people whose responsibilities include engineering, production, account- ing and plant management. The intention is that theory about changes in the role of cost accounting will emerge from an analytical synthesis of the hypotheses and supporting findings. 24For additional sources on process-oriented research see George and McKeown (1985), and Mohr (1985). 95 DESIGN GUIDELINES FOR THE FIELD STUDY Five sources, Bonoma (1985), Kaplan (1985b), Mohr (1982, 1985), Stone (1978), and Yin (1981, 1984), were relied upon to establish speci- fic design guidelines for the on-site investigation procedures. Several guidelines were derived collectively from these sources: a Site Selection. Several sites were visited to enable access to multiple sources of evidence, thus helping to reveal emerging patterns. Sites were selected to enable comparison of attri- butes expected to be similar and unaffected by change, and contrast of attributes expected to be different and affected by change. I Focus of Inquiry. To make sure the inquiry procedures had focus and purpose, along with reasonable consistency of procedure at all research sites, each field visit was governed by an orderly plan and a field interview guide was prepared and used. The guide was prepared in draft form for the first site visit and then modified for the other visits. a Seeking Richness in Data. A variety of means of data gathering were used at the sites, including interviews, documents, sketches, etc., to capture some of the richness of the context within which the subject of inquiry is operating. a Data Gathering Procedures. Care was taken that the research procedures did not intrude upon the subject of inquiry, causing the outcome to be manipulated by the inquiry procedures. Multi- ple sources of data input were tapped at each site to provide some degree of verification. This helped to increase the chan- ces of discovering explanations through emerging patterns. Each of these guidelines were considered and implemented in the study design, as discussed in the sections that follow. SW The plant sites selected for the study are metal fabricators in Michigan and nearby states. Elements of CIM have been implemented in half of the sites. The firms produce a variety of products in large 96 volumes. The products and manufacturing processes are similar enough to be able to use a comparative field study design. Potential research sites were identified by obtaining recommenda- tions from faculty members, CIM consultants, and participants at profes- sional conferences on cost accounting and CIM, and by identification of companies publicized in business journals and newspapers. Telephone calls were made to high level executives of potential companies to explain the purpose of the study and to request an initial interview. Following the initial interviews with executives of participating com- panies, an arrangement letter was sent to confirm the intent and general arrangements for their participation. An example of an arrangement letter is reproduced in Appendix II. The initial intention was to identify a single set of sites com- prised of manufacturers in the same industry, half of which would be CIM plants. Such a set of plants could be differentiated on the CIM dimen- sion and compared on other dimensions, and their industry characteris- tics and manufacturing process characteristics would be reasonably com- parable. However, early discussions with potential firms quickly revealed strong sensitivity toward competitors. The firms were uncom- fortable with the notion of research being conducted within their busi- ness and also with their competitors. An alternative approach was adopted, which was to seek pairs of sites, with similar products for each pair, but with one of the pair a CIM producer and the other a traditional producer. Four pairs of manu- facturers of metal products were selected. In two cases, two plants out 97 of divisions of the same firm were paired. The pairings of firms selected for the study were: Pair 1 Producers of steel sheet metal products Pair 2 Producers of components for automobiles Pair 3* Producers of components for trucks Pair 4* Producers of frames for vehicle passenger seats * Pairs within the same company In the case of each pair, one of the sites had elements of CIM already implemented or in the process of being implemented. WW Certain criteria were used to classify half of the sites as CIM sites. The initial intention was to classify a site as a CIM site if it had "MRP II" attributes. "MRP II” refers to "Manufacturing Resources Planning,” a broader connotation than "MRP," an acronym for "Materials Requirements Planning' [Wight (1981)]. Wight (1981) provided classi- fication criteria useful for determining how closely specific firms fit the ideal of an MRP II firm. In particular, Wight's ideal MRP II firm has a "whole company” system with three features of MRP II (Wight (1981): a The accounting system and manufacturing operating system are integrated, using the same transactions and the same measurement data. The accounting data are extended from operating data. a The MRP II data systems have simulation capability, making it possible to handle "what if” inquiries. a The MRP II system is a whole company system, involving all busi- ness facets that pertain to manufacturing resources: sales, production, purchasing, inventory, planning and coordinating of schedules, and cash flow. Initially, the intention was to classify a site "CIM” if opera- tional data systems had a "whole company" perspective and cost and 98 production data were used to simulate alternatives. These criteria were used to classify four of the firms as CIM sites when selecting them for site visitation. Upon visiting the sites, the CIM sites all appeared to meet the first and third of the above criteria, but none satisfied the “what if” simulation criterion. While all of the CIM sites were capable of using their integrated data systems for "what if" simulations, they were not doing so to any substantial degree. Ultimately, the classification decision was resolved by applying as additional criteria certain elements of CIM derived from literature reviewed in Chapter III: I Use of Materials Requirements Planning (MRP) in planning resource requirements I Use of manufacturing cells I Presence of "pull-through" control of production flow a Adoption of specific procedures to reduce set-up times I Use of computer-aided design (CAD) and computer-aided machinery (CAM) and robotics equipment. Specific characteristics of the each selected site are summarized later in the chapter in Table 7. WWW Field studies are designed to ensure that the field inquiry is focused. Yin (1984, p. 31) says the focus of inquiry can be considered the "unit of analysis," for which the effects of change are being ex- plored. For this study, the unit of analysis is the role of cost ac- counting (as defined) in a context of change as CIM is introduced. An interview guide was used to help ensure that the scope and subject 99 matter of inquiry stayed on track and in focus. Both interviews and documentation were the sources of empirical findings for determination of the role of cost accounting. MW The primary means of collecting empirical field data was discussion with individuals who hold production, engineering or accounting posi- tions. Based on Stone (1978) and Yin (1984), the following objectives were adopted for interviews and other data gathered in the field: a During interviews, listening to the interviewee was responsive and yet non-obtrusive. Affirmative (but neutral) replies were used to encourage elaboration. a Descriptive terms, acronyms or measures expressed by interview- ees were followed with requests for definition and examples. Terms defined in one way at one site were occasionally defined differently at others. a Examples, both in the form of oral scenarios and documents, were requested. Copies of documents were annotated to describe and define headings, captions, and terms. a To a certain extent, questions were repeated across individuals at the same sites to strengthen the validity of findings. However, statements made by one interviewee were not revealed to other interviewees. Each interviewee was assured that their specific comments were taken in confidence and if quoted, their identity would not be revealed. a Interviews were taped with the permission of the interviewee (no secret recordings were made). a Maps or sketches were obtained for plants toured and were annotated to show resources and resource activities. a With permission of management, photographs were taken of some areas toured. Assurance was given that the photographs would not be published or shown to others and all proprietary informa- tion would be protected. Each of the objectives was incorporated into the field data gather- ing procedures in the study. All interviewees were willing to have 100 their interviews taped. A limited number of photographs were obtained, but were not shown to others and are not included in the study. To summarize, design objectives for the field study included site selection, focus of inquiry, multiple sources and types of data, and responsive, unobtrusive interviewing without intervention on the views and knowledge of the interviewees. FIELD RESEARCH PROCEDURES This section describes four aspects of the field research proced- ures. First, criteria used to select four pairs of field sites are described. Table 5 shows the contrasts sought in choosing and pairing the sites. Then, using hypothetical names, the sites actually selected are listed in Table 6 and described in some detail in the ensuing pages. Table 7 summarizes site characteristic described in the discussion. Finally, a field interview guide is described, which contains a set of nine manufacturing decisions (listed in Table 8) used to focus the site interviews. Specific site visit procedures are described in the final part of this section. CRITERIA FOR SITE SELECTION The main objective was to find sites with general similarity in the nature of their products manufactured and their manufacturing processes. For CIM sites: (1) the sites would be similar with regard to use of computer- aided technology and integrated manufacturing processes, 101 (2) data would be available to interpret how computer integration had changed the manufacturing activities, (3) contrasts between the CIM firms and the traditional manufac- turers were to concentrate on changes made to CIM firms' cost accounting systems, and (4) ideally, the sites were to provide comparative insights about changes in the quality and availability of cost information after the introduction of CIM. -- v Four pairs of firms were selected for the study. Each pair had similar production activities and output. One site had implemented CIM, and the other was a traditional manufacturer. Table 5 summarizes the desired characteristics of the pairs of firms. TABLE 5 COMPARISONS OF INDIVIDUAL PAIRS OF FIRMS 119.111 QIMj Traditionalk Products Similar Similar Manufacturing Processes Similar Similar Manufacturing Control Systems CIM Traditional Cost Accounting Adapted Traditional Pairi [i - 1,2,3,4] are the CIM/Traditional pairs listed in Table 6; CIM [j - A,C,E,G] are CIM firms; and Traditionalk [k - B,D,F,H] are tra itional firms. As Table 5 shows, this field study was carried out in several manufacturing company locations where (1) the manufacturing processes 102 are devoted to the fabricating or assembly of components and are rela- tively similar across the locations, (2) production can be carried out using either CIM or traditional manufacturing processes, and (3) cost accounting data can be observed and analyzed. The design called for field work at four pairs of manufacturing sites, including pairs of divisions or plants within individual firms. CIM has been introduced at four of the sites to a sufficient degree that comparison of the cost accounting activities and use of accounting data is possible. The comparisons addressed two dimensions: (1) across the entire set of sites and (2) within the CIM firms, comparing the present and past manufacturing controls and cost accounting systems. The bases of com- parison are those suggested by the research hypotheses and supporting focus points detailed in Chapter III. THE SITES DESCRIBED Eight plant sites were selected where products are manufactured out of metal. Some of the plants manufacture heavy products from steel castings or forgings; others manufacture products from sheet steel. By arrangement with the participating companies, the identities of the sites are not revealed and hypothetical names are used. Table 6 shows the pairings of sites visited. 103 TABLE 6 PAIRINGS OF SITES VISITED CIM TRADITIONAL _$_IIE.E___S.IIE§__ Pair 1: Abnett Bladnu Pair 2: Ceston Dolnar Pair 3: Elnep Flaxtin Pair 4: Gledbul Holpin Each of the plants is described in the next several pages, followed by Table 7, which summarizes comparative data about the plants. Abnett -- (91M §l§§l Abnett plant produces a major component for large trucks. The product is machined and assembled from steel forgings and bar stock. Manufacturing processes include machining, grinding, welding, cutting, heat treating and assembly. The cost of the product is comprised of 55 percent materials, 7 percent labor and 38 percent overhead. Overhead costs at the plant are about 550 percent of labor costs. Production volume is approximately 7,000 units per month, although in the past pro- duction has been as high as 25,000 units per month. The plant facility was built in 1969, is 360,000 sq. ft. in size, and has about 320 produc- tion employees. The "Abnett” and ”Bladnu" plants (Site 2) are both part of a vehi- cle components division of a large multi-divisional manufacturing com- pany. Division management recommended that both Abnett and Bladnu 104 plants be visited, Abnett as a CIM site and Bladnu as a traditional site. Since customers were imposing strict expectations concerning timing and sequence of delivery, restructuring was underway in the Abnett plant to make it the focus plant for final assembly. Changes were being implemented at Abnett to create a pull-through production flow pattern based on very precise scheduling of final assembly. A substantial capital investment had been made in a flexible machining center (which was being installed) adjacent to the assembly area. This large multi-task machining center would manufacture parts for two new models to feed to assembly, where the machined parts would be combined with other components supplied from the Bladnu plant, another plant of the division. Abnett had moved three machining lines over to the Bladnu plant to make room for the flexible machining center. A major emphasis on just-in-time (JIT) pull-through production flow was also taking place. The aim was to achieve reliable delivery of assembled product with uninterrupted production flow, while tolerating variety in the product models. As stated by the Plant Manager: "Timing is everything! Flow is everything!“ In addition, major reductions in set-up times had been achieved by changing set-up procedures. Abnett was going through a transition from a traditional setting to a CIM setting, re-organizing the patterns of production flow to derive a pull-through Just-in-Time flow rather than batch production. Cells were being formed and set-up times were being reduced. A substantial amount of investment had been made in computer-aided machinery. For these reasons, Abnett was classified as a CIM site in the study. 105 “mm! -- (IIEQIEIQDBJ $132) Bladnu is another plant in the same division as Abnett, but located in a different city. Bladnu's plant was built in 1972, occupies 490,000 sq. ft, and has about 500 employees, including 350 direct laborers. There are seventeen production cost centers in the plant. Bladnu pro- duces machined component parts used in the product assembled at Abnett. Manufacturing processes include cutting and grinding of rough forgings, and heat treating. Cost content of the product is 52 percent materials, 7 percent labor and 40 percent overhead. The plant overhead rate is nearly 600 percent of direct labor. About $8 million of inventory is in the plant. Production operates on a push-through basis, scheduled by an MRP system. The MRP system is not a closed-loop and does not provide formal order releases. There are daily meetings of foremen to make arrange- ments to cover shortages and immediate production order commitments. There are ten numeric control machines in use at the plant, al- though no significant investments in CIM technology have been made in the past three years. Much of the production equipment is inflexible, and dedicated. However, Bladnu has been conducting a study to determine how to change the processes to permit using more flexible automation. The Bladnu plant does not have elements of CIM or JIT and therefore has been classified as a traditional plant for the study. QEEEQD .- (91! $152) Ceston and Dolnar (Site 4) are two plants in a division of a large company that manufactures a component of automobiles and small trucks. 106 Ceston was recommended by division management as a CIM site and Dolnar as a traditional site. Ceston's plant is about 280,000 sq. ft. in size, was built in 1966 and employs 400 people, including 350 hourly workers. The plant produces fifty models of its end product. Manufacturing processes include stamping, tube making, wire bending, welding, and some limited assembly. There are no sub-assemblies. Cost content of the product is 50 percent materials, 10 percent labor and 40 percent over- head. The plant overhead rate is approximately 400 percent of direct labor. The plant was originally laid out in a process orientation, with functional departments. Recently, the production flow has been rear- ranged to achieve a pull-through pattern oriented to product families. Nearly $100,000 was spent to rearrange equipment into cells. The rea- sons were to reduce production throughput time, do a better job of meet- ing schedule commitments, improve quality, and utilize more capacity without buffer inventory. Inventory has been cut from $4 million to less than $2 million. Scrap rates have also been reduced. Nine robots are used in the plant. Visual information is provided in prominent places in the plant to inform employees about the schedule. 'Ranban' cards are used in the stampings area to trigger replenishment of stampings as used.25 Container sizes limit quantities of inventory in production areas. Cells have been established, where improved labor utilization and reduced tooling and rework costs have been observed. 25"Kan‘ban" are move cards that authorize replenishment of con- tainers when emptied. See Hall (1983, ch. 3) for further discussion and an example. 107 CAD is used for new product designs. There are about ten to fifteen new designs each year. Standard costs are used, and are revised annually. Ceston has been classified in the study as a CIM plant because of the rearrangements made in the plant to implement JIT concepts and achieve pull-through production flow. DQIDII -- (IIHQIEIQDEJ Slflfil Like the Ceston plant, the Dolnar plant manufactures welded steel seat frames for use in motor vehicles. The plant is 250,000 sq. ft. in size, has about 300 employees (160 direct laborers) and $1.4 million of inventory. Purchased materials for the products include steel tubing and thick steel wire, both of which are cut, shaped and welded to make seat frame assemblies. Manufacturing processes include stamping, wire bending and welding. There are several models used for seat cushions and seat backs for motor vehicle manufacturers. The products are mature in design and have about a three-year design life cycle. The cost content of the products is 57 percent materials, 8 percent labor and 35 percent overhead. The average departmental overhead rate is about 450 percent of direct labor. Dolnar's plant was in the early stages of reorganizing its produc- tion flow patterns to implement pull-through JIT concepts. The plant had proposed a $700,000 capital expenditure for costs of plant rearran- gements and flexible machining centers. The objectives of the proposed expenditure were to expand its daily capacity and improve its on-time delivery performance. The plant was near new plants being constructed by potential major customers. Some cells had already been created in 108 the plant to focus on certain models, including cells where part of the equipment has been provided by the customer. Four robotic machines were in use and more were planned. Recent capital investments included about $250,000 for information systems enhancements to support pull-through production under JIT concepts. The Dolnar plant was aiming to accomplish a change-over to a JIT pull-through production environment, just as its sister plant, Ceston, had already done. At the time of the visit, however, production was essentially batch-oriented, and scheduled on a push-through basis by an MRP system. Since the Dolnar plant lacked the integrated processes of a CIM facility, it has been classified as a traditional plant. HIDE! -- (GIN $152) The Elnep plant is a 330,000 sq. ft. facility built during the 19608. The plant has approximately 350 production employees (260 direct workers). A major component for large trucks is manufactured at the plant. The product is mature, has a highly respected quality repu- tation, and supplies a major share of the U.S. market for the product. Procured raw materials, which comprise 40 percent of the product's costs, include castings, forgings, bar stock, valves and other miscel- laneous hardware. Manufacturing processes include turning, grinding, drilling and other heavy metal machining operations, followed by heat treating, finishing and assembly. Direct labor represents about 12 percent of product costs, and Manufacturing Overhead 48 percent. Prac- tical capacity of the plant is 250 units per day. Actual volume at the time of the visit was about 130 units per day. 109 Elnep's plant is one of three plants in a division of a large multi-divisional U.S. company. Of the three plants, division management recommended the Elnep plant as the one with the most CIM attributes: use of CNC machines, CAD/CAM, MRP, and manufacturing cells. Prior to the visit, the plant had made significant reductions in inventories and improvements in inventory turnover.26 Cells had been formed in several areas to reduce production throughput time and reduce costs by taking advantage of commonality of parts across models. One divisional execu- tive responsible for materials management said: The plants have reduced their throughput time by about 50 percent. Today it takes five days to process a gear and seven days to process a shaft. Four years ago it took four weeks for a gear and four to six weeks for a shaft. Elnep has been classified as a CIM plant for the study, although some characteristics blur the classification. Elnep's MRP system is not a "closed loop" system because capacity planning and shop floor control are not integrated with MRP scheduling. Furthermore, although cells are used in several areas the overall pattern of production flow is not a pull-through just-in-time (JIT) pattern. However, Elnep was undertaking a program to implement CIM. The program had been developed by divis- ional management, involved the investment of several hundred thousand dollars in equipment and rearrangement costs and was underway at the time of the visit. On balance, classifying the Elnep plant as a CIM plant seemed to be the most appropriate classification. 26Schonberger (1986) cited the plant as an "honor roll” plant because of its substantial reductions in inventory and improvement of inventory turnover. 110 EJIEEID -- (IIEQIEIQDBJ gjgg) The Flaxtin plant is owned by a large manufacturer of motor vehi- cles. The Flaxtin plant manufactures a motor vehicle component out of sheet metal. The 123,000 sq. ft. site (with approximately 125 employ- ees) is part of a larger plant area in which other components are also manufactured. A real-time accounting data access system was in use on an experimental basis in the manufacturing process area. Data obtained at the site pertained to both the site and to the larger plant opera- tions of which the site is a part. Flaxtin is not a CIM site, since the equipment, process lay-out and other aspects of production are tradi- tional. Thus, the Flaxtin site was classified as a traditional site in the study. Elfiflhfll -- (Q15 51:21 Gledbul is a large (800,000 sq. ft. with 1,100 employees) plant facility, which manufactures an assembled sheet metal product for com- mercial use. Raw materials include coil steel, bar stock, wheels and ornamental materials. Machining operations include stamping, cutting, grinding, welding, and assembly. Cost content of the product is about 45 percent materials, 6 percent labor and 49 percent overhead. The average overall overhead rate is about 800 percent of direct labor. Gledbul is owned by a leading manufacturer of commercial products made from sheet metal, a company which has a national reputation for innovative products and superior quality. Its plants use advanced technology to produce with very high reliability. Gledbul uses CAD/CAM extensively, has numerous manufacturing cells, robotics, manufacturing 111 cells and an effective ”closed loop" MRP system. Inventory turns over very rapidly at the plant; about five days of production output is in the plant at any one time (amount of inventory was not made available). The Gledbul plant was classified as a CIM plant because its operations are well integrated, rely on very accurate information and use advanced CIM technology, including manufacturing cells, pull-through production, computer-aided design linked with computer-aided processes, and minimal inventory. -- a te The Holpin plant is a 180,000 sq. ft. traditional plant with four functional process areas: initial fabrication, final fabrication, paint- ing and final assembly. The plant is a subsidiary company owned by a large manufacturer of a commercial sheet metal product. Eighty percent of production is sold to the parent company. Holpin has approximately 130 employees, including 60 direct laborers. Processes include cut- ting, roll forming and stamping of coil steel, and welding. Cost con- tent of the product is 48 percent materials, 9 percent labor, and 43 percent overhead. The average plantwide overhead rate is 515 percent of direct labor. Production is batch oriented. Standard costs are not used for product costing or monthly transac- tions. An MRP system is being implemented for the first time, and steps are underway to orient employees to MRP and to improve data accuracy. Holpin lacks CIM attributes and therefore is classified as a traditional plant in the study. 112 ect Having described characteristics of each site in some detail, Table 7 summarizes the selected sites, with the CIM sites on the first page of the table and traditional sites on the second. 1113 >.aouceaa x x x x x x x x .aoo. no.0.u. x ammoccu-dmaa moon mc_uc_sa use mc.m.es .mc.c.:osa .mc_mtoe use uc_uu:o .auem as» o.e_couea .eucoaecto .a.oegs .xooua can .Leeuo ._ou woe no nae uoamota .e.oteslou >..e:cc< x x XXX zusotgu.:aaa Nooo >dcmoaae mc_:a_c_m .mc.ueecu use: .mceaasu .mc....cu .mc_vc_tm .mc.ccsp 00> CLOCCUIIV .ao>.s> .xoouu can .amc_mto~ .umc.uasu ace umw Roe amuse» tow ucocoasou >..e:cc< x x XXXXX X sonata“...=a Room uc_u.o: .oc_ucos ot_: .uc_xsa one» 0: ucmnsu .05.: .oe«« new new mom no.u_go> toe ucocoaaou ssgucox x x XXXXXX zoaoazu-..:a acmm >dnaoaas .mcmudos .mc.mc_cm .mc_c_soea .uc.uu:u ¢0> moons can .omc.mcom .eeuu Run as 8mm acocoaaoo musty con-m «Now odnedme>e uoz 000' 000’ oo—.F can can ON» ooo.ooo ooc.onn ooo.ou~ ooo.oon damawao mummy mobuwu hhmmm< a h z < 4 a x _ u aubua owhuumwu no uuuhm—cwhu<¢<:u no >¢et eo secescotm auaoo vases-um a...» ueuetmouc_ nup:m—mph< mm_pm:ouu< ucoan_:oe o_uono¢ noc.goea uzu .. :(u oO aooaoooca mo aoa>p no.hu_-mhu<¢h useeto>o con-4 ade.couem "acoucoo «sou co_ua.coeea mu_pu_uwbu¢m..s3cc< x x gusocgu.gosa edaa._s>s uoz >.oseaaa .mcmumes .uc_a2aua a.» o_uua.a .Lsuoa uses» emce._a>s uoz o.ne._a>a uoz omne._s>e no: acocoaaoo Laue: uoocm >..e:cc< x x goaocgu.:a:a Rome mc_mmes .mc.mcen 05.: .mc.aesua 0: mcmcsu .et_s .oeum man no as“ ao.o_:o> toe ucocoaaou >.guco: x x XXX guaocgu.gusa mace mcmumos .mc_uaocu .mc.ueotu use: .mc_mc_cm .mc_c.guea .mc.uu:u 0: moons can .amc_mtom now as awn Auuocn< to». eucecoaaou o.n.._.>. so: can....>. so; «one. >.tou ~ao. as. m~. m¢~ can ooo.oow ooc.n~_ ooa.om~ coo.ooe a_asoa a.»xoc we mocoacoam nunoo vases-um ao._w countenan— mmhamnmhb< om—hmsouu< ucosamaoo o_uoco¢ seemsoea uxu .. ao neaaeooca eo seam» mu—pa—uupuo tones a.e_touex "acoucoo unou co_ua.aoeoa mu_ha_muhu.tsog we .oz .ue .c. "on.“ mm—»_4_uo Hausa momw>wn-nsm no; woman on .Ho>oso: .muouoom ocoaou oomsomou so momma muonsoo once 0» hauoouwn mumoo oceano>o mo annex ouoa wcawumso one: mouan 2H0 .muowwwuu soauoo -msouu once as use: no: one: consumes souuoso -ouo mmoooua-euusH .uuowmfiuu sowuoomaeuu mom muaaoo moo momma use musuoo someone» uaxo\zuuso to mass Aassoauuusuu one 2H0 muonv «some HH< .vo>uomno now one? souuoowmwucoow umoo o>oumma ou mcoauoowmamnsao uadooos no nmouaoo umoo Ga newcono ocean .ooufiomom mo cacao he condom» no: one: monsoooo .Ho>oao= .mooudomou Hsoawoa -oanoou mo owned: one woodman ou monoganmumo one: musdoooo boa oaom .mHHoo noauou mass: .muouoom condemou uco>oaou so muses -cmammm umoo omen ou use voumonmon condomou mo mmuHo hp mumoo uncomom vouuoomsm haxsoz awamme ou mouoomxo ones woman 2H0 Hamflmonm< Hmou .mcowuoomssuu umoo Howmuuu ou muso>o udouommHo .Hnaoauaons nouuomean uoz can on nouooexo one: moves zHu a .sowuoOHMauaoua condomou mo mmsao he muouaoo umoo noouw ou mouoonxo one: use unoucoo unoo AuoHHoau many ouoa new haze eon .oomuom one: muoucoo once so: oaom nouuooosm masses o>m£ ou nonsense one: mouam zHo o H 82 o 90ng 38% gnaw a qmzmck mquZHomwx mom Hmommbm mo Mudxzpm Ha Handy 203 .usouxo unsoamstam s on muse umoo ou mmooom oGHH-:o on: was mam mouwm 2H0 och .oaoum oauu Bouuhn cowuosvouo one Sousa uo: man maoumha umoo esp mo madam oaau may .osuom use henna .wsHHoou .maoauouoa so muuooon .mo>wuoonno wsauauoomscna ouwououun haxoos no Adamo omoeoum ou ouosumoa uoonmnsoumm do once muuoaou umoo mason one means ones mucoussooos uaoam .aoumho umoo .ousu once on mucous o:«~-:o omen ecu scum wcwuuoaou awaken a“ no: use mousnoooun om: .haucosooum once upon umoo smoumhu-mm03 Hoamomcn aw oo>uomno one: newcssu nouuooesm thsoB ounces ou nouoooxo owes woman 2H0 gag mg g g3 u<=mgm as; I- I‘l‘l‘hh" EC >n