Quality cost measurement under activity-based costing Wen-Hsien Tsai National Central University, Chung-Li, Taiwan, Republic of China Introduction Many companies in the world gradually promote quality as the central customer value and regard it as a key concept of company strategy in order to achieve the competitive edge (Ross and Wegman, 1990). Measuring and reporting the cost of quality (COQ) is the first step in a quality management program. Even in service industries, COQ systems receive considerable attention (Bohan and Horney, 1991; Carr, 1992; Ravitz, 1991).
COQ systems are bound to increase in importance because COQ-related activities consume as much as 25 percent or more of the resources used in companies (Ravitz, 1991). COQ information can be used to indicate major opportunities for corrective action and to provide incentives for quality improvement. Traditional cost accounting, whose main functions are inventory valuation and income determination for external financial reporting, does not yield the COQ information needed.
While most COQ measurement methods are activity/process oriented, traditional cost accounting establishes cost accounts by the categories of expenses, instead of activities. Under traditional cost accounting, many COQ-related costs are lumped into overheads, which are allocated to cost centers (usually departments) and then to products through predetermined overhead rates. For example, among various COQ-related costs, the rework and the unrecovered cost of spoiled goods caused by internal failures are charged to the factory overhead control account which accumulates the actual overhead costs incurred (Hammer et al. 1993, pp. 155-64). The predetermined overhead rates should be adjusted to incorporate the normal levels of various COQ-related costs, and excess COQ-related costs will be buried in overhead variances. The cost accounting treatment described above cannot satisfy the needs of COQ measurement. Thus, Oakland (1993, p. 210) claims that “quality related costs should be collected and reported separately and not absorbed into a variety of overheads”. Prevention-appraisal-failure (PAF) approach and process cost approach are two main approaches to measuring COQ.
However, The author wishes to thank the anonymous referees for many helpful comments and suggestions about this paper. The author also wishes to thank the authors of references cited in this paper, especially the authors, Barrie G. Dale and James J. Plunkett, of the book, Quality Costing, and the author, Peter B. B. Turney, of the book, Common Cents: The ABC Performance Breakthrough – How to Succeed with Activity-based Costing, from which this paper quotes a lot of COQ and BC concept. Quality cost measurement 719 Received March 1996 Revised March 1998 International Journal of Quality & Reliability Management, Vol. 5 No. 7, 1998, pp. 719-752, © MCB University Press, 0265-671X IJQRM 15,7 720 these approaches still cannot provide appropriate methods to include overhead costs in COQ systems. Accordingly, many quality cost elements require estimates and there is a prevailing belief in COQ literature. It is a danger that managers become too concerned with accuracy in COQ determination – a number-crunching exercise that will consume resources disproportionately (Oakland, 1993, p. 197). In addition, most COQ measurement systems in use do not trace quality costs to their sources (O’Guin, 1991, p. 0), which hinders managers from identifying where the quality improvement opportunities lie. Nevertheless, these deficiencies could be easily overcome under activity-based costing (ABC) developed by Cooper and Kaplan of Harvard Business School (Cooper, 1988; Cooper and Kaplan, 1988). ABC uses the two-stage procedure to achieve the accurate costs of various cost objects (such as departments, products, customers, and channels), tracing resource costs (including overhead costs) to activities, and then tracing the costs of activities to cost objects.
The purpose of this paper is to present a conceptual framework for measuring quality costs under ABC. First, the present approaches to measuring COQ are reviewed; second, the two-dimensional model of ABC and activitybased management (ABM) is explained; third, COQ approaches and ABC are compared and an integrated COQ-ABC framework is presented; fourth, COQ measurement, COQ reporting, and uses of COQ information under ABC are discussed; finally, a hypothetically simplified example is presented to illustrate how to measure COQ under ABC.
Approaches to measuring COQ Since Juran (1951) discussed the cost of quality, many researchers have proposed various approaches to measuring COQ. Reviews of COQ literature can be found in Plunkett and Dale (1987) and Porter and Rayner (1992). In this section, we will briefly review the approaches to measuring COQ. PAF approach After Feigenbaum (1956) categorized quality costs into prevention-appraisalfailure (PAF), the PAF scheme has been almost universally accepted for quality costing.
The failure costs in this scheme can be further classified into two subcategories: internal failure and external failure costs. Oakland (1993, pp. 186-9) describes these costs as follows: • Prevention costs: These costs are associated with the design, implementation and maintenance of the total quality management system. Prevention costs are planned and are incurred before actual operation. • Appraisal costs: These costs are associated with the supplier’s and customer’s evaluation of purchased materials, processes, intermediates, products and services to assure conformance with the specified requirements.
Internal failure costs: These costs occur when the results of work fail to reach designed quality standards and are detected before transfer to customer takes place. • External failure costs: These costs occur when products or services fail to reach design quality standards but are not detected until after transfer to the customer. Quality cost elements. In order to calculate total quality cost, the quality cost elements should be identified under the categories of prevention, appraisal, internal failure and external failure costs.
BS 6143: Part 2 (1990) and ASQC (1974) have identified a list of quality cost elements under this categorization. These lists just act as a guideline for quality costing. Most elements in these lists are not relevant to a particular industry, while many elements identified by practitioners are peculiar to an industry, or a company (Dale and Plunkett, 1991, p. 28). Some typical COQ elements are shown in Table I. In the initial stages of the quality costing exercise, some companies put mphasis on just identifying the costs of failure and appraisal activities. The methodology usually used is for each department, using a team approach, to identify COQ elements which are appropriate to them and for which they have ownership. Several techniques, such as brainstorming, nominal group technique, Pareto analysis, cause and effect analysis, fishbone diagrams, and forcefield analysis, can be used to effectively identify COQ elements (Dale and Plunkett, 1991, p. 41; Johnson, 1995).
The quality cost measurement system developed will improve with use and experience and gradually include all quality cost elements. Economics of quality-related activities. One of the goals of total quality management (TQM) is to meet the customer’s requirements with lower cost. For this goal, we have to know the interactions between quality-related activities associated with prevention, appraisal, internal failure and external failure costs. It will help in finding the best resource allocation among various quality-related activities.
In the literature, there are many notional models describing the relationships between the major categories of quality costs. Generally speaking, the basic suppositions of these notional models are “that investment in prevention and appraisal activities will bring handsome rewards from reduced failure costs, and that further investment in prevention activities will show profits from reduced appraisal costs” (Plunkett and Dale, 1988). Plunkett and Dale (1988) classify these notional models into five groups, which are further aggregated into three by Burgess (1996).
After a critical review, Plunkett and Dale (1988) conclude that “many of the models are inaccurate and misleading, and serious doubts are cast on the concept of an optimal quality level corresponding to a minimum point on the total quality-cost curve”. Besides, Schneiderman (1986) asserts that, in some circumstances, if enough effort is put into prevention, no defects at all would be produced, resulting in zero failure costs and no need for appraisal (also given in Porter and Rayner (1992)). Thus, in these circumstances, the only optimal point is “zero-defects”. • Quality cost measurement 721 IJQRM 15,7 Categories Prevention
COQ elements Quality control and process control engineering Design and develop control equipment Quality planning by others Production equipment for quality – maintenance and calibration Test and inspection equipment – maintenance and calibration Supplier quality assurance Training Administration, audit, improvement Laboratory acceptance testing Inspection and test In-process inspection (non-inspectors) Set-up for inspection and test Inspection and test materials Product quality audits Review of test and inspection data On-site performance testing Internal testing and release Evaluation of materials and spares Data processing, inspection and test reports Scrap Rework and repair Troubleshooting, defect analysis Reinspect, retest Scrap and rework: fault of supplier Modification permits and concessions Downgrading Complaints Product service: liability Products returned or recalled Returned material repair Warranty replacement Loss of customer goodwill a Loss of sales a 722 Appraisal Internal failure External failure Table I. Typical COQ elements Note: a Intangible external failure costs (not included in BS 6143: Part 2) Source: BS 6143: Part 2 (1990) (also given in Dale and Plunkett (1991, p. 71))
However, Burgess (1996) integrated the three types of quality-cost models, derived from reducing Plunkett and Dale’s categories (1988), into a system dynamic quality-cost model displaying dynamic behavior consistent with published empirical data. According to the simulation results, Burgess concludes that the simulation provides support for the classic view of qualitycost behavior that an optimal level of quality exists only in certain timeconstrained situations. If the time horizon is infinite, or above a particular cutoff value, then spending on prevention can always be justified, i. e. the modern view prevails. Drawbacks of the PAF approach.
Although the PAF model is universally accepted for quality costing, there are a number of criticisms of it (Oakland, 1993, pp. 200-1; Porter and Rayner, 1992), described as follows: • It is difficult to decide which activities stand for prevention of quality failures since almost everything a well-managed company does has something to do with preventing quality problems. • There are a range of prevention activities in any company which are integral to ensuring quality but may never be included in the report of quality costs. • Practical experience indicates that firms which have achieved notable reductions in quality costs do not always seem to have greatly increased their expenditure on prevention. Original PAF model does not include intangible quality costs such as “loss of customer goodwill” and “loss of sales”. • It is sometimes difficult to uniquely classify costs (e. g. design reviews) into prevention, appraisal, internal failure, or external failure costs. • The PAF model focuses attention on cost reduction and ignores the positive contribution to price and sales volume by improved quality. • As mentioned above, the classic view of an optimal quality level is not in accordance with the continuous quality improvement philosophy of TQM. • The key focus of TQM is on process improvement, while the PAF categorization scheme does not consider process costs. Therefore, the PAF model is of limited use in a TQM program. Alternatives to the PAF approach.
The alternatives to the PAF categorization scheme are divisions of quality costs into conformance and nonconformance, tangible and intangible, controllable and uncontrollable, discretionary and consequential costs, and so on. Crosby (1984, p. 86) divides quality costs into two categories: (1) the price of conformance (POC), including the explicitly quality-related costs incurred in making certain that things are done right the first time; and (2) the price of nonconformance (PNOC), including all the costs incurred because quality is not right the first time. Crosby’s POC includes prevention and inspection costs and his PNOC includes internal and external failure costs (Shank and Govindarajan, 1994, p. 6).
In Xerox, quality costs are classified into three categories: (1) the cost of conformance (prevention and appraisal); (2) the cost of nonconformance (failure to meet customer requirements before and after delivery); and (3) the cost of lost opportunities (Carr, 1992). Quality cost measurement 723 IJQRM 15,7 724 The first two categories are analogous to Crosby’s POC and PNOC respectively. Xerox measures lost opportunities as the profit impact of lost revenues. It occurs when a customer chooses a competitive product over a Xerox product, when a customer cancels an order because of inadequate service, or when a customer buys Xerox equipment that is inadequate or unnecessary and switches to another brand. Juran claims that both prevention and appraisal costs are inevitable and are not worth including in quality costs (Juran et al. , 1975).
Juran advocates a categorization of quality costs including: • • • tangible factory costs, which are measurable costs such as scrap, rework, and additional inspection; tangible sales costs, which are measurable costs such as handling customer complaints and warranty costs; intangible costs, which can only be estimated, such as loss of customer goodwill, delays caused by stoppages and rework, and loss of morale among staff (also given in Porter and Rayner (1992)). Juran’s categorization scheme focusses on the costs of product failures and emphasizes the importance of intangible quality cost elements, which in the long term are of greater importance than cost reduction. Another alternative, proposed by Dale and Plunkett, is to consider the activities relating to supplier, company (in-house) and customer under the PAF categorization. This approach has the merit of new categories which closely relate to the business activities while retaining the advantages of the PAF categorization (Dale and Plunkett, 1991, pp. 26-7).
Process cost approach It seems that the identification of quality cost elements in the PAF approach is somewhat arbitrary. It may focus on some quality-related activities which account for the significant part of total quality cost, not on all the interrelated activities in a process. Under the philosophy of process improvement in TQM, analysts should place emphasis on the cost of each process rather than an arbitrarily defined cost of quality (Goulden and Rawlins, 1995). In view of this, the process cost approach, described in the revised BS 6143: Part 1 (1992), is proposed. It recognizes the importance of process cost measurement and ownership.
The process cost is the total of the cost of conformance (COC) and the cost of nonconformance (CONC) for a particular process. The COC is the actual process cost of providing products or services to the required standards, first time and every time, by a given specified process. The CONC is the failure cost associated with a process not being operated to the required standard (Porter and Rayner, 1992). According to this definition, we know that the content of this categorization (COC and CONC) is different from that of Crosby’s (POC and PONC) and Xerox’s (COC and CONC) mentioned previously. The process cost model can be developed for any process within an organization.
It will identify all the activities and parameters within the process to be monitored by flowcharting the process. Then, the flowcharted activities are allocated as COC or CONC, and the cost of quality at each stage (i. e. COC + CONC) are calculated or estimated. Finally, key areas for process improvement are identified and improved by investing in prevention activities and process redesign to reduce the CONC and the excessive COC respectively (Oakland, 1993, pp. 201-7; Porter and Rayner, 1992). The British Standards Institution has included this methodology in the revised BS 6143: Part 1 (1992). Dale and Plunkett (1991, p. 43) state that this will help to extend the concept of quality costing to all functions of an nterprise and to non-manufacturing organizations, and that it also gets people to consider in more detail the processes being carried out within the organization. A process modelling method, IDEF (the computer-aided manufacturing integrated program definition methodology) (Ross, 1977), can be used to construct the process cost models for the processes within an organization (Marsh, 1989). This method utilizes activity boxes with inputs, outputs, controls and mechanisms to depict the activities of a process. However, the IDEF method is developed for use by experts for system modeling. It seems to be too complex if departmental manager and staff were to be responsible for identifying the elements of process costs.
Thus, Crossfield and Dale (1990) develop a more simple method called quality management activity planning (Q-MAP) for the mapping of quality assurance procedures, information, flows and qualityrelated responsibilities. In addition, Goulden and Rawlins (1995) propose a hybrid model to overcome the limitations of the IDEF method for process quality costing. It constructs integrated or functional flowcharts to represent the processes by using information from a three-level hierarchical model (function-department-activity), where the lowest level is an activity defined as a task with a single output. This will reduce the level of complexity of the IDEF type models. Other COQ approaches There are some approaches to measuring COQ in addition to the PAF approach and the process cost approach.
For example, Son and Hsu (1991) proposed a quantitative approach to measuring quality costs, which considers both manufacturing processes and statistical quality control. Statistical terms of quality are translated to dollar terms in this approach. However, the quality cost model presented in their paper is restricted to a simplified manufacturing system which consists of only a machining area (with in-process sampling inspection) and a final inspection area (with 100 percent final inspection). Besides, Chen and Tang (1992) present a pictorial approach to measuring COQ, which is patterned after that used in a computer-based information system design.
The COQ variables considered in this approach include direct COQ variables (PAF costs and quality-related equipment costs) and indirect Quality cost measurement 725 IJQRM 15,7 726 COQ variables (customer-incurred costs, customer-dissatisfaction costs and loss of reputation). It includes two major steps: (1) specifying the COQ variables as well as the significant relationships among the variables, and mapping the variables and relationships into an “influence diagram” showing the structure of a COQ system; and (2) converting the structure into a well-defined “entity-relationship diagram” showing the input-output functions and their associated properties.
The influence diagram used in the pictorial approach can provide an easy-tounderstand COQ system for quality management practitioners, and the entityrelationship diagram can provide an effective framework for maintaining and modifying the COQ system. Quality cost collection Fundamentally, the PAF approach, the process cost approach and even Chen and Tang’s pictorial approach need first to identify the quality cost elements. For the PAF approach, COQ elements are identified under the cost categories of the selected categorization scheme. Most of COQ elements relate to qualityrelated activities. For the process cost approach, the cost elements of COC and CONC for a process are derived from the flowcharted activities of the process.
Most cost elements of COC do not relate to quality-related activities of traditional COQ view. After identifying the quality cost elements, we should quantify the elements and then put costs (dollar values) on the elements which have been identified (Dale and Plunkett, 1991, pp. 40-1). In COQ literature, many authors pay attention to why COQ information is important and what should be included in a COQ system, and seldom discuss how to measure and collect quality costs. However, Dale and Plunkett give lots of guidance on quality cost collection, including objectives and scope, approaches, sources of data, ease of collection, level of detail, accuracy of data, and people involved (Dale and Plunkett, 1991, Ch. 3, pp. 6-51). Besides, Thorne (1990) recommends some relatively uncomplicated methods for calculating COQ, i. e. collecting quality costs by account, by defect type, by whole person, by labor hours, and by personal log (also given in Johnson (1995)). Dale and Plunkett state that “the collection and synthesis of quality costs is very much a matter of searching and shifting through data which have been gathered for other purposes” (Dale and Plunkett, 1991, p. 38). Some of quality costs are readily available from a cost accounting system (e. g. scrap and rework costs); some can be derived from the data of activity reports (e. g. repair and inspection costs).
Nevertheless, a large portion of quality costs should be estimated by some ways. For example, the opportunity costs of lost customer goodwill and lost sales, which are intangible external failure costs (similar to the third category of Xerox’s COQ system), can be estimated by Taguchi’s quality loss function (Albright and Roth, 1994). Other examples are the costs of producing excess inventories and material handling due to suboptimal plant layouts, which are indirect failure costs and can be estimated by expertise. In addition, calculating prevention costs needs the estimates of apportionment of time by indirect workers and staff who do not usually record how they spend their time in various activities.
Deficiencies of most COQ systems Generally speaking, there are the following deficiencies in measuring COQ among most COQ systems: • The aspect of overhead allocation in calculating COQ is seldom discussed in the literature. In practice, some companies add overheads to the direct cost of labor and material on rework and scrap, while other companies do not. If they do, “rework and scrap costs become grossly inflated compared with prevention and appraisal costs which are incurred via salaried and indirect workers” (Dale and Plunkett, 1991, p. 45). • Most of COQ measurement systems in use are not (there are some exceptions) intended to trace quality costs to their sources (O’Guin, 1991, p. 70) such as parts, products, designs, processes, departments, vendors, distribution channels, territories, and so on.
Accordingly, the COQ information derived from these systems cannot be used to identify where the quality improvement opportunities exist. • “It is the general lack of information about how people, other than direct workers, spend their time which presents a considerable obstacle to the collections of quality costs” (Dale and Plunkett, 1991, p. 112). The deficiencies mentioned above will decrease the accuracy of quality costs and limit the usefulness of a COQ system. However, these deficiencies can be easily solved under ABC (Cooper, 1988; Cooper and Kaplan, 1988), developed by Cooper and Kaplan of Harvard Business School, together with other techniques.
Two-dimensional model of ABC Evolution of ABC The main shortcoming of traditional cost accounting is to distribute overhead costs over products by using volume-related allocation bases such as direct labor hours, direct labor costs, direct material costs, machine hours, etc. It will not seriously distort the product cost in the conventional manufacturing environment where overheads are just a small portion of product cost. In the modern manufacturing environment, however, the overheads will grow rapidly as manufacturers increasingly promote the level of automation and computerization, and the cost distortion of traditional cost accounting will be significant (Brimson, 1991, p. 179).
The main reason is that many overhead costs vary with volume diversity, product diversity, and volume-unrelated activities (e. g. set-up and scheduling activities), not with the volume-related measures. For example, highvolume products consume more direct labor hours than low-volume products, but Quality cost measurement 727 IJQRM 15,7 728 high-volume products do not necessarily consume more scheduling cost than lowvolume products. Therefore, traditional cost accounting will overcost highvolume products and undercost low-volume products. In view of this, Cooper and Kaplan suggested using ABC to improve the accuracy of product costs. In early ABC systems (Turney, 1991, pp. 7-80), overhead cost is divided into various cost pools, where each cost pool contains the cost of a group of related activities consumed by products in approximately the same way. Each cost pool is distributed to products by using a unique factor that approximates the consumption of cost. This unique factor, called an allocation basis in traditional cost accounting, could be volume-related (e. g. direct labor hours and machine hours) or volume-unrelated (e. g. number of orders, set-up hours, and number of parts). Early ABC systems focus on the accurate assignment of overhead costs to products. They do not provide direct information about activities and do not consider the costs outside the plant.
Thus, a two-dimensional model of ABC is proposed as shown in Figure 1. This ABC model is composed of two dimensions: cost assignment view and process view described in the following two subsections. The cost assignment view of ABC The detailed cost assignment view of ABC is shown in Figure 2. ABC assumes that cost objects (e. g. products, product lines, processes, customers, channels, Cost Assignment View Resources Resource Drivers Process View Cost Drivers Activities Performance Measures Activity Drivers Cost Objects Figure 1. Two-dimensional model of ABC Source: Adapted from Turney, (1991, p. 81) Resources Resource Drivers First Stage Quality cost measurement Activity Center
Activity & Activity Cost Pool 729 Cost Element Second Stage Activity Drivers Cost Objects Figure 2. Detailed cost assignment view of ABC Source: Adapted from Turney, (1991, p. 97) markets, etc. ) create the need for activities, and activities create the need for resources. Accordingly, ABC uses two-stage procedure to assign resource costs to cost objects. In the first stage, resource costs are assigned to various activities by using resource drivers. Resource drivers are the factors chosen to approximate the consumption of resources by the activities. Each type of resource traced to an activity becomes a cost element of an activity cost pool.
Thus, an activity cost pool is the total cost associated with an activity. An activity center is composed of related activities, usually clustered by function or process. We can create activity centers by various ways according to different information needs. In the second stage, each activity cost pool is distributed to cost objects by using an adequate activity driver which is used to measure the consumption of activities by cost objects (Turney, 1992). If the cost objects are products, then total cost of a specific product can be calculated by adding the costs of various activities assigned to that product. The unit cost of the product is achieved by dividing the total cost by the quantity of the product.
The resources used in manufacturing companies may include people, machines, facilities, and utilities, and the corresponding resource costs could be assigned to activities in the first stage of cost assignment by using the resource drivers time, machine hours, square footage, and kilowatt hours respectively (Brimson, 1991, p. 135). For the manufacturing activities, there are the following categories of activities (Cooper, 1990): • unit-level activities (performed one time for one unit of product, e. g. 100 percent inspection, machining, finishing); IJQRM 15,7 730 batch-level activities (performed one time for a batch of products, e. g. sampling inspection, set-up, scheduling); • product-level activities (performed to benefit all units of a particular product, e. g. roduct design, design verification and review); • facility-level activities (performed to sustain the manufacturing facility, e. g. plant guard and management, zero defect program). The costs of different levels of activities will be traced to products by using the different kinds of activity drivers in the second stage of ABC cost assignment view. For example, machine hours is used as the activity driver for the activity machining; set-up hours or the number of set-up for machine set-up; and the number of drawings for product design (Tsai, 1996b). Usually, the costs of facility-level activities cannot be traced to products with the definite causeeffect relationships and should be allocated to products with appropriate allocation bases (Cooper, 1990).
The information achieved from ABC cost assignment view is usually used for the decisions of pricing, quoting, product mix (Tsai, 1994), make versus buy, sourcing, product design, profitability analysis, and so on (Turney, 1992). The process view of ABC The process view of ABC is composed of three building blocks: cost drivers, activities and performance measures. It provides information on why the activities are performed via cost drivers and on how well the activities are performed via performance measures. Cost drivers are factors that determine the workload and effort required to perform an activity (Turney, 1991, p. 87), i. e. factors that cause a change in the cost of an activity (Raffish and Turney, 1991). For example, the quality of parts received by an activity (e. . the percent that are defective) is a determining factor in the work required by that activity, because the quality of parts received affects the resources required to perform the activity. Cost drivers identify the cause of activity cost and are useful because they point people to take action at the root cause level, i. e. they reveal opportunities for improvement. An activity may have multiple cost drivers associated with it. Performance measures are used to indicate the work performed and the results achieved in an activity (Raffish and Turney, 1991). They tell us how the activity is meeting the needs of its internal or external customers.
There are five fundamental elements of activity performance: (1) the quality of the work done; (2) the productivity for the activity; (3) the cycle time required to complete the activity; (4) the cost traced or allocated to the activity; and (5) customer satisfaction (Miller, 1996, pp. 94-5; Turney, 1991, pp. 88-9). • Performance measures differ from one activity to another and from one company to another. Performance measures may be financial or nonfinancial. The process view of ABC places emphasis on processes. A process is a series of activities that are linked to perform a specific objective. A business process often runs across artificial organizational boundaries, departments or functions. Because of the interdependency of activities in a process, the work of each activity affects the performance of the next activity in the process.
Accordingly, performance measures for one activity may become cost drivers for the next activity (Turney, 1991, p. 190). For example, performance measures for designing new tools may include the number of change in specifications and the number of new drawings, and these performance measures are just the cost drivers for the succeeding activity, manufacturing new tools. This tells us that merely identifying activities without consideration of their relationship to other activities in the process will result in overlooking many improvement opportunities (Lawson, 1994, p. 35). The information achieved from the process view of ABC can be used to aid in process/activity improvement.
The potential improvement opportunities can be located by performance measurement and value analysis. First, the areas where the improvement is needed can be identified by comparing this period’s performance with performance goals or with best practices of comparable activities inside or outside the company. The latter comparison is called benchmarking (Turney, 1991, p. 111). Second, the areas where the improvement is needed can be identified through the categorization of activities as valueadded or non-value-added. An activity is value-added if it is judged to contribute to customer value or satisfy an organizational need; otherwise, it is non-valueadded.
For the non-value-added activities, improvement initiatives should be directed toward eliminating or minimizing the activities. For the valued-added activities, improvement initiatives should be directed toward streamlining, improving what is being done, and optimizing performance (Miller, 1996, pp. 923). However, the potential improvement opportunities should be prioritized by Pareto analysis. We can rank the activities in descending order of cost achieved from the cost assignment view of ABC, and determine the significant activities that will provide the greatest opportunities for improvement. Usually, we will find that 20 percent of the activities cause 80 percent of the cost. Thus, these significant activities are worth improving in the first place.
After the top-priority areas of improvement are recognized, cost driver analysis can be used to examine, quantify, and explain the effects of cost drivers of the significant activities mentioned above. This will help direct improvement efforts to the cause of cost and avoid treating the symptom (Miller, 1996, p. 93). For example, inspecting the incoming material is a non-value-added activity and its cost driver will be the quality of material received from suppliers. If we have sufficient confidence in the quality of material received from suppliers, we may conduct sampling inspections, even no inspections for incoming material. Otherwise, we may need 100 percent inspections. Therefore, the best way to reduce the efforts of incoming material inspections is to choose the suppliers Quality cost measurement 731 IJQRM 15,7 hat provide the high-quality material or to help our suppliers establish the quality control/assurance systems. Activity-based management Using ABC to improve a business is called activity-based management (ABM). As defined by the consortium for Advanced Manufacturing International (CAM-I) (Raffish and Turney, 1991), ABM is a discipline that focuses on the management of activities as the route to improving the value received by the customer and the profit achieved by providing this value. This discipline includes cost driver analysis, activity analysis, and performance measurement. ABM uses the cost and nonfinancial/operational information acquired from ABC in various analyses.
For example, ABM uses the cost information of activities, products, customers, and other cost objects, supplied by the cost assignment view of ABC, to perform strategic decision analysis (such as pricing, product mix, sourcing, customer profitability analysis), activity-based budgeting, life-cycle costing and target costing. In addition, ABM uses the information provided by the process view of ABC to support cost reduction, downsizing, process/quality improvement, benchmarking, business process reengineering (BPR), and total quality management (TQM). This paper focuses on COQ measurement and continuous quality improvement under ABC. Comparison between COQ approaches and ABC The PAF approach of COQ is activity-oriented, the process cost approach of COQ is process-oriented, and ABC is activity-oriented for the cost assignment view and process-oriented for the process view.
Accordingly, the PAF approach of COQ regards COQ-related (or PAF-related) activities as improvement objects, and the process cost approach of COQ and ABC regard processes/activities as improvement objects. A summary comparison between COQ approaches and ABC is given in Table II. As for activity/cost categories, COQ approaches and ABC have the following classifications: • PAF approach – prevention, appraisal, internal failure, and external failure; • process cost approach – conformance and nonconformance; • ABC – value-added and non-value-added. Some authors classify prevention and appraisal costs as the cost of conformance, and internal and external failure costs as the cost of nonconformance.
However, the content and meaning of conformance and nonconformance costs of the PAF approach are different in nature from that of the process cost approach. Under the ABC perspective, only prevention costs in the PAF approach (Ostrenga, 1991) and only some of conformance costs in the process cost approach are value-added, and their relationships are shown in Figure 3. Thus, the cost of nonconformance either in the PAF or in the process 732 COQ Aspects of comparison Orientation Activity/cost categories PAF approach Activity-oriented Prevention Appraisal Internal failure External failure Process cost approach ABC Quality cost measurement Process-oriented Activity-oriented (cost assignment view) Process-oriented (process view) Conformance Value-added Nonconformance Non-value-added 733
Treatment of overhead No consensus method to allocate overhead to COQ elements under current COQ measurement systems and traditional cost accounting Assigning overhead to activities by using resource drivers in the first stage of ABC cost assignment view Tracing activity costs to cost objects by using activity drivers in the second stage of ABC cost assignment view Processes/activities Process/activity value analysis Performance measurement Benchmarking Cost driver analysis Tracing costs No adequate method to trace to their sources? quality costs to their sources Improvement objects Tools for improvement COQ-related activities Processes activities
Quality circle Brainstorming Nominal group technique Cause and effect analysis Fishbone diagram Forcefield analysis The cost elements of PAF categories Total quality cost and the costs of PAF categories/ elements and their percentages of various bases The COC and CONC elements of the processes investigated Total process cost, COC, and CONC of the processes investigated and their percentages of various bases Information outputs The costs of activities and processes The costs of value-added and non-valueadded activities and their percentages of various bases Accurate costs of various cost objects (e. g. product, departments, customers and channels) Activity-based performance measures Cost drivers of activities Table II. Comparison between COQ approaches and ABC Related management technique TQM ABM PAF approach Prevention Appraisal Internal Failure External Failure ABC Value-added Non-value-added Process cost approach Conformance Nonconformance Figure 3. The relationship between activity/cost categories of COQ approaches and ABC IJQRM 15,7 734 ost approach would be eliminated or minimized through investment in prevention activities. The cost of conformance in the process cost approach would be reduced by streamlining or redesigning the process. As for information outputs, the fundamental cost information outputs achieved from the PAF approach are the costs of PAF-related activities; from the process cost approach and ABC are the costs of activities and processes. While all these three methods will provide the costs of activities, ABC will create a variety of information outputs. From the discussion above, we can see that there are many similarities in process perspective between the process cost approach and ABC.
In addition, the similarities between these two methods also can be found in the steps of process improvement by the process cost approach and by ABC/ABM, as shown in Table III. In this Table, the corresponding steps are in the same rows, where we will find that these two methods deal with the same thing by using different terminology. Integrated COQ-ABC framework From the explanation in the previous sections, we know that ABC can supply various cost and nonfinancial information to support COQ programs. Moreover, ABC can provide more accurate costs of activities and processes than traditional cost accounting, which make COQ information more valuable for TQM. Hence, it is better to integrate COQ approaches with ABC. Figure 4 shows an integrated COQ-ABC framework.
In this framework, we may adopt the PAF approach or the process cost approach for COQ measurement. Strictly speaking, COQ-related activities for the PAF approach or flowcharted activities for the process cost approach should be incorporated into the building block “activities” of ABC model. That is, ideally, ABC and COQ blocks in this framework should be merged as one. Furthermore, there are the following characteristics in the integrated COQ-ABC framework: • ABC and COQ systems should share the common database in order to avoid data redundancy and inconsistency. • The related management techniques of ABC and COQ are ABM and TQM, respectively. ABC system can provide cost and activity/process-related information for ABM, COQ/TQM, and business process reengineering (BPR). • BPR is another management technique for process improvement. The major difference between BPR and ABM/TQM is that BPR overthrows and improves current business processes in a fundamental and radical way. This can be seen from the definition of BPR: “BPR is the fundamental rethinking and radical redesign of business processes to achieve dramatic improvements in critical, contemporary measures of performance, such as cost, quality, service, and speed” (Hammer and Champy, 1993). Process cost approach 1. Choose a key process to be analyzed 2. Define the process and its boundaries 3.
Construct the process diagram: (1) Identify the outputs and customers (2) Identify the inputs and suppliers (3) Identify the controls and resources 4. Flowchart the process and identify the process owners. The process owners form the improvement team 5. Allocate the activities as COC or CONC 6. Calculate or estimate the quality costs (COC + CONC) at each stage. Estimates may be required where the accounting system is unable to generate the necessary information 7. Construct a process cost report 8. Prioritize the failure costs and select the process stages for improvement through reduction in costs of nonconformance (CONC). This should indicate any requirements for investment in prevention activities.
An excess cost of conformance (COC) may suggest the need for process redesign. 9. Review the flowchart to identify the scope for reductions in the cost of conformance. Attempts to reduce COC require a thorough process understanding, and a second flowchart of what the new process should be may help 10. Monitor conformance and nonconformance costs on a regular basis, using the model and review for further improvements (i. e. continuous improvement) ABC/ABM 1. Define critical business processes and specify key and significant activities 2. Select process owners 3. Define the process boundaries 4. Form and train process improvement teams 5. Flowchart the processes 6.
Analyze activities: (1) Define activity outputs/measures (2) Identify the customer/user of activity outputs (3) Perform value-added analysis (4) Identify cost drivers (5) Determine activity performance measures and goals (6) Define other activity attributes: • Primary versus secondary activities • Core, sustaining, and discretionary • Cost behavior: fixed or variable, direct or indirect, avoidable or unavoidable (7) Gather activity data required for activity/cost object costing and for activity/process improvement 7. Perform activity/process cost assignment 8. Summarize processes and costs for management 9. Draw up the process improvement plan: (1) Locate the potential improvement opportunities by performance measurement and value analysis (2) Prioritize the improvement opportunities by Pareto analysis and select the significant activities that will provide the greatest opportunities for improvement (3) Design and select the improvement alternatives by cost driver analysis, process redesign, new process design (process innovation), and others 10.
Implement the process improvement plan 11. Monitor the process improvement results by installing performance measurement and feedback control systems to ensure that the desired results are achieved and to provide feedback for continuous improvement Quality cost measurement 735 Sources: 1. Steps of process/quality improvement by the process cost approach come from Oakland (1993, pp. 201-3). 2. Steps of process improvement by ABC/ABM are synthesized from Beischel (1990); Harrington (1993); Miller (1996, pp. 69-98); O’Guin (1991, pp. 306-13); Ostrenga and Probst (1992). Table III. Steps of process improvement by the process cost approach and ABC/ABM IJQRM 15,7
Missions To profitably improve the value received by the customer Objectives To promote productivity To eliminate waste To reduce throughput time To reduce cost To improve quality To increase the equity owned by the shareholder 736 Goal Continuous Process/Activity/Quality Improvement Tools Cost Driver Analysis Forcefield Analysis Fishbone Diagram Process/Activity Value Analysis Performance Measurement Benchmarking Quality Circle Business Process Reengineering Cause and Effect Analysis Nominal Group Technique Brainstorming A B M T Q M Cost Assignment View Resources PAF Approach COQ-related Activities Prevention Process View Resource Drivers Appraisal Internal Failure External Failure Cost Drivers Activities Performance Measures
OR Process Cost Approach Activity Drivers Cost Objects Flowcharted Activities of Processes COC CONC Figure 4. Integrated COQ-ABC framework Activity-Based Costing Common Database COQ Approach • The common goal of ABM, TQM, and BPR is continuous improvement by using various respective tools. In fact, all these tools can be used in ABM, TQM, and BPR. The common objectives of ABM, TQM, and BPR are to promote productivity, to eliminate waste, to reduce throughput time, to reduce cost, and to improve quality. • • The ultimate missions are to profitably improve the value received by the customer and to increase the equity owned by the shareholder. Quality cost measurement
COQ measurement under ABC As mentioned before, there is no consensus method to allocate overheads to COQ elements and no adequate method to trace quality costs to their sources under current COQ measurement systems and traditional cost accounting. These deficiencies can be overcome easily by using the cost assignment view of ABC (as shown in Figure 1), either in the PAF approach or in the process cost approach. Overhead costs are traced to activities by using resource drivers in the first stage of ABC cost assignment view. Then, activity costs are traced to their sources (i. e. cost objects) by using activity drivers in the second stage of ABC cost assignment view.
COQ measurement For the PAF approach, the activities of the ABC model would be COQ-related activities (prevention, appraisal, internal failure, and external failure) and COQunrelated activities. In the first stage of ABC cost assignment view, resource costs (including overhead costs) of the company are traced to various COQunrelated and COQ-related activities (as shown in Table I) by using resource drivers. The resources used by COQ-related activities may be people, computers, equipment, material (parts), supplies, facilities, energy, and so on. If a resource is dedicated to a single COQ-related activity, so the resource cost is directly traced to that COQ-related activity.
If a resource supports several COQrelated and/or COQ-unrelated activities, the resource cost must be distributed among these activities by using a appropriate resource driver. The resource driver of people-related costs (salaries and benefits) will be time. If a COQrelated activity uses a worker’s partial time, this COQ-related activity will receive the worker’s salary and benefit according to its usage percentage of the worker’s time. For another example, the resource driver of energy costs will be kilowatt hours. The cost of any specific COQ-related activity will be achieved by adding all the resource costs (i. e. activity cost elements) traced to that COQrelated activity.
Therefore, each of the four components of total COQ can be obtained respectively by accumulating the costs of all the activities related to that COQ component. Finally, total COQ is the sum of the four components’ costs. Accordingly, total COQ, four COQ components, and the costs of detailed COQ-related activities can be achieved from the first stage of ABC cost assignment view. For the process cost approach, the activities of the ABC model would be the flowcharted activities of various processes, including COC-related and CONC-related activities. The method of tracing resource costs (including overhead costs) to activities is the same as described above.
The results achieved in the first stage of ABC cost assignment view would be the costs of flowcharted activities, total process costs, COC, and CONC of various processes. 737 IJQRM 15,7 738 The treatment of overheads in ABC is different from the practice of including full overhead costs in direct labor charges to quality-related costs. It will produce accurate quality costs and solve the usual problem in current COQ systems: double-counting. Another deficiency, mentioned before, of a combination of current COQ systems and traditional cost accounting is the lack of information about how indirect workers, whose costs are one part of overhead costs, spend their time on various activities.
This deficiency make prevention the most difficult of the categories to cost because it depends heavily on the estimates of percentage of time spent by indirect workers and staff (Dale and Plunkett, 1991, p. 44). In practice, these estimates are often derived from information gathered from interviews or questionnaires (Turney, 1991, p. 277). Nevertheless, the information acquired from interviews and questionnaires may be under suspicion because “the way that people actually spend their time, after all, can often be quite different than the way they think that they spend it” (Miller, 1992). To overcome this deficiency, Tsai (1996a) suggested using work sampling to estimate the percentage of the indirect worker time spent on each activity.
Work sampling (Richardson, 1976), which utilizes the random sampling techniques, allows one to understand the characteristics of a process by collecting data on portions of a process rather than the entire process. This technique is particularly useful in the analysis of nonrepetitive or irregularly occurring activities. With the aid of computer software (Lund, 1990), it is feasible to use work sampling to provide more accurate data of first stage resource drivers for indirect workers in the integrated COQ-ABC systems. Tracing COQ to its sources We can trace COQ to its sources (such as parts, products, designs, processes, department, vendors, distribution channel, territories, etc. ) through the second stage of ABC cost assignment view.
If we, under the PAF approach, need to know the COQ information by departments or products, then we could regard departments or products as cost objects and trace the various COQ-related activity costs to departments or products by using appropriate activity drivers in the second stage of ABC cost assignment view. Most activities related to prevention costs are sustaining activities and their costs are not easy to be traced to departments or products. It is because of no explicit cause-and-effect relationship between prevention activities and departments (or products). On the other hand, cause-and-effect relationships do exist between departments (or products) and most activities related to appraisal, internal failure, and external failure costs, so these costs could be traced to departments or products appropriately. Similarly, we could trace external failure costs to distribution channels or territories (used as the cost objects).
For the process cost approach, the elemental information achieved from the first stage of ABC cost assignment view are the COC and CONC elements of flowcharted activities of processes. For the purpose of this paper, we want to trace the cost of CONC-related activities to their sources. This also can be achieved by using appropriate activity drivers in the second stage of ABC cost assignment view. COQ reporting under ABC From the above discussion, we know COQ reports for the PAF approach can be easily prepared under ABC in the following ways: • COQ reports associated with detailed COQ-related activities for the whole company. • COQ reports associated with activities related to appraisal, internal failure, external failure costs by departments, products, or product lines. COQ reports associated with activities related to external failure costs by distribution channels or territories. These are just some illustrative COQ reports. These COQ reports usually provide monthly costs, year-to-date costs, variances to budgeted costs, and a comparison with the previous years’ cost data. These COQ reports may include the COQ percentage of various bases such as sales revenue, manufacturing cost, units of product, and so on (Simpson and Muthler, 1987). In addition, trend analysis can be used to compare present COQ data with historical COQ data in order to know how COQ changes over time. In order to prepare the above COQ reports, we could use activity centers to group the related activities under ABC.
There may be hundreds of activities in a company. Activity centers allow us to easily locate activities with identical characteristics. If we group the company’s activities by processes, then we will have the COQ activity center and several other activity centers. Within the COQ activity center, four nested activity centers can be established, i. e. prevention, appraisal, internal failure, and external failure activity centers (as shown in Figure 5). Again, we could, within a sub-activity center, set up a nested activity center on demand. This will create hierarchies of COQ information and give us the different levels and breadths of COQ information.
This will form a multi-tier COQ reporting system. We could use attributes, which are labels describing the type of activity, to create activity centers on demand in the integrated COQABC information system (Turney, 1991, pp. 271-3). If the COQ activity center is divided according to the company’s organizational structure, then we could, by adopting the concept of responsible accounting, provide department managers COQ Quality cost measurement 739 Preventing Appraisal Internal Failure External Failure Figure 5. COQ activity center and its nested activity centers IJQRM 15,7 740 with the COQ information and related nonfinancial information which are specific to them.
The COQ responsibility reports will help responsible managers identify the major costs of nonconformance incurred in their departments and actuate improvement projects to eliminate or reduce these costs. For the process cost approach, the COQ reporting method is the same as described above except that PAF-related activities are replaced with COCrelated and CONC-related activities. Uses of COQ information Under the integrated COQ-ABC framework, the quality system and the ABC system must be integrated to produce COQ information and the related operational information of activities and processes. The information achieved can be used in various aspects (Dale and Plunkett, 1991, pp. 59-68; O’Guin, 1991, pp. 71-5). Some of the important uses of COQ information are described as follows.
To identify the magnitude of the quality improvement opportunities ABC, together with other techniques such as work sampling, can trace resource costs (including overhead costs) to various activities in a rational way which avoids double-counting. Thus, ABC can create the accurate costs of PAF-related activities for the PAF approach and of COC- and CONC-related activities for the process cost approach. The prime purpose of the quality improvement is to gradually eliminate or reduce the cost of poor quality, i. e. to improve the activities related to the appraisal and failure costs for the PAF approach and to the CONC for the process cost approach.
ABC will tell management the accurate cost of poor quality and indicate which activities are the most expensive through Pareto analysis. Accordingly, management can identify the directions and magnitude of the quality improvement opportunities. This information is useful in the investment justification of the quality improvement alternatives such as investment in prevention activities or equipment. To identify where the quality improvement opportunities exist By integrating the ABC system and the quality system, the cost of poor quality can be traced to its sources. Hence, the integrated system can identify where the quality improvement opportunities exist and provide the following benefits (O’Guin, 1991, p. 2): • By tracing quality losses to product attributes, parts, processes, engineering, and vendors, management can take corrective actions toward the right direction. • By tracing and costing vendor returns by vendor and parts, purchasing managers will understand the true costs of buying from particular vendors. This will avoid forcing purchasing managers to buy strictly on price. If scrap costs result from worker errors, the scrap costs are assigned to the process’s overheads. This will provide management with clear pictures of who is causing defects and how much they cost. • The integrated system arms the quality department with defect and rework cost information.
Some defects are more costly than others and some mean much more to the customers than others. Thus, the system can tell the quality department where to concentrate its quality improvement efforts. • By tracing warranty and return costs to their products, it will eliminate the tendency for product managers to rush a product through testing, or ship defective goods to achieve their sales targets. Before tracing quality costs to their sources, we should dig out their root causes by using the cost driver analysis of ABC process view in order to direct improvement efforts to the cause of cost and avoid treating the symptom. Table IV gives a list of some possible cost drivers of four COQ components.
For example, the root causes of the internal failure rework could be design error, defective purchased material, deficient tooling and maintenance, and worker error. If we find out the real root cause of excessive rework cost is defective purchased material, then we could effectively solve the problem by helping improve the supplier’s quality system or searching another supplier. For another example, if excessive in-process inspections are due to complex design, then we can encourage designers to simplify the design by using the number of different part numbers as a performance measure or activity driver. To plan the quality improvement programs A quality improvement program should depict quality improvement actions, improvement targets, and quality cost budgets.
Improvement targets may be set COQ components Prevention Appraisal Cost drivers Investment in reducing overall COQ-related activities Set-up frequency Tight tolerance activities Complex design Design error Defective purchased material Machine reliability Tooling age and conditions Worker error Order entry errors Incorrect assembly instructions Product failure Worker error • Quality cost measurement 741 Internal failure External failure Source: Adapted from Ostrenga (1991, p. 43) Table IV. Some cost drivers of COQ components IJQRM 15,7 742 after quality improvement actions are evaluated and selected. Under this circumstance, improvement targets are set and quality cost budgets are prepared according to the savings of required activity driver quantities for each quality-activity of the selected quality improvement actions.
On the contrary, quality improvement actions may be worked out according to improvement targets set by management just like the approach of target costing. In this scenario, management may establish quality improvement targets for every unit of the organization. Management may request purchasing to reduce vendor quality costs by 20 or 40 percent in a year. The amount of rework, which is the activity driver of rework activity, may be targeted to be cut by 30 percent. Target quality costs are derived from the quality improvement targets under the present operational conditions. Then, various quality improvement actions are evaluated one by one till the ones, whose budgeted quality costs are not greater than target quality costs, are found.
In either case, quality cost budgets are constructed incorporating improvement targets by using the budgeted activity driver quantities based on improved activities and the moving average rates (or the last period’s rate) of activity drivers. This method can be applied in either the PAF approach or the process cost approach (Sharman, 1996). To control quality costs Since management establishes quality improvement targets for every unit of the organization, management can then track actual performance to these targets after one period’s operation. If improvement targets are not met, the variances between actual and budgeted quality costs will emerge. The variances may represent unanticipated quality loss. The variances force management to exploit what is causing the variance and encourage management to eliminate the source of quality loss.
This feedback allows management to continuously plan the quality improvement programs and control quality costs (O’Guin, 1991, p. 74). The method described above is the budget control of quality costs. It may report quality costs monthly. However, it is not fit for daily operation control. Turney (1991, pp. 197-9) demonstrated how ABC was used for total quality control by utilizing daily COQ reporting in a printed circuit board (PCB) plant. The ABC system was used to prepare a report on the cost of poor quality for each activity immediately after each of the three daily shifts and to show graphically the trend in physical defects and cost. The report allows management to focus immediately on the quality problems with the biggest cost impact.
The ABC system also prepared daily a top ten offenders list that reports the ten products with the highest cost of poor quality on the previous day. It pinpointed the poor quality products and provided the greatest potential for redemption. When a product unit on this list was scrapped, a report, which showed the cause of the problem as well as the cost, was prepared and sent to the person most likely to correct the problem. This use of ABC made quality problems visible within a matter of hours, or even minutes. Turney suggested linking ABC with computer-integrated manufacturing (CIM) to prepare COQ reports for real-time and cost-effective control. An illustration of COQ measurement under ABC As an illustration, a hypothetically simplified example is presented in this section.
A department in a manufacturing company produces two products, Product A and Product B. This company adopts the PAF approach to measure COQ. The workers in this department carry out nine distinct activities. The related information of these activities is shown in Table V, including required resources, activity levels, PAF categories, value-added or non-value-added, and activity drivers. Some of these activities are PAF-related activities, and some are not. There are six workers in this department. The hourly labor cost for each worker is $10 calculated on the basis of wages and benefits. Four of the workers carry out five activities, machining, rework, warranty repair, inspection, and package.
These workers spend most of their time in direct work, and they will record how they spend the time by using time cards. The other two workers carry out four activities, scheduling, maintenance, set-up, and material handling, which are indirect works. Work sampling is used to estimate the percentages of time spent on these four activities for two indirect workers. For the machining activity, assume that Product A and Product B can be processed in two general-purpose machines. The hourly machine cost for each machine is $20 calculated on the basis of the costs of machines and their tools. The activities, rework and warranty repair, are also carried out in these two machines.
For the inspection activity, Product A needs two tests, while Product B needs only one. In addition, assume that there are 20 work days in a specific month with eight work hours per day and that the production quantities of Product A and Product B are 225 and 350 respectively in the month. Now, the department manager wants to calculate COQ and unit product costs. Quality cost measurement 743 Activities Machining Rework Warranty repair Inspection Required resources People, machines, tools People, machines, tools People, machines, tools Activity levels Unit – – Value-added (VA) PAF or non-valuecategories added (NVA) – Internal failure External failure Appraisal – – Prevention – – VA NVA NVA NVA VA Gray a VA NVA NVA
Activity drivers Machine hours # of reworks # of warranty repairs # of tests # of units # of batches Machine hours # of set-ups # of moves People, test equipment, Unit tools, supplies Package People, tools, supplies Unit Scheduling People Batch Maintenance People, supplies Facility Set-up People, tools Batch Material People, moving Batch handling equipment Note: a Gray: Activities of no value to customers, but that may be essential to the functioning of the department Table V. Activity-related information for the illustration IJQRM 15,7 744 Resource cost assignment In the first stage of ABC cost assignment view, resources are traced to activities.
For the labor resource, the labor hours of the workers spent on various activities and their percentages are shown in Table VI. Note that the labor hours of the worker #1 to #4 spent on the first five activities are acquired from time cards and that the percentages of time of the workers #5 and #6 spent on the last four activities are estimated from a work sampling study (Tsai, 1996a). Total labor costs traced to activities are shown in the last column of Table VI. For the machine resource, the machine hours of the machines used in the activities, machining, rework, and warranty repair, are recorded and shown in Table VII. Total machine costs traced to activities are shown in the last column of Table VII.
For the other resources consumed by activities, their costs can be directly traced to activities. Therefore, various resource costs and total activity costs traced to activities are shown in columns (1)-(4) of Table VIII. From column (4) of Table VIII, the department manager knows that the four high-cost activities are machining ($6,416), inspection ($2,088), set-up ($1,920), and Package ($1,840). Inspection and set-up are non-value-added activities, Activities Machining Rework Warranty repair Inspection Package Scheduling Maintenance Set-up Material handling Idle Table VI. Assigning labor resource costs to activities Total #1 99. 2 (62) 24. 0 (15) 16. 0 (10) 8. 0 (5) #2 100. 8 (63) 16. (10) 19. 2 (12) Workers #3 #4 #5 #6 Total Total labor labor hours $/hour costs 200. 0 (20. 8) 40. 0 (4. 2) 51. 2 (5. 3) 139. 2 (14. 5) 153. 6 (16. 0) 20. 8 (2. 2) 44. 8 (4. 75) 160. 0 (16. 7) 67. 2 (7. 0) 83. 2 (8. 7) 960. 0 (100) 10 10 10 10 10 10 10 10 10 10 10 $2,000 (20. 8) 400 (4. 2) 512 (5. 3) 1,392 (14. 5) 1,536 (16. 0) 208 (2. 2) 448 (4. 7) 1,600 (16. 7) 672 (7. 0) 832 (8. 7) $9,600 (100) 9. 6 (6) 8. 0 (5) 120. 0 (75) 16. 0 (10) 8. 0 (5) 11. 2 (7) 128. 0 (80) 8. 0 (5) 12. 8 (8) 96. 0 (60) 32. 0 (20) 11. 2 (7) 160. 0 (100) 12. 8 (8) 32. 0 (20) 64. 0 (40) 35. 2 (22) 16. 0 (10) 160. 0 (100) 12. 8 (8) 160. 0 (100) 14. 4 (9) 160. 0 (100) 6. 0 (10) 160. 0 (100) 12. 8 (8) 160. 0 (100) Note: Figures in parentheses are labor hours indicated as a percentage Machines Activities Machining Rework Warranty repair Idle Total #1 108. 8 (68) 20. 8 (13) 14. 4 (9) 16. 0 (10) 160. 0 (100) #2 112. 0 (70) 14. 4 (9) 16. 0 (10) 17. 6 (11) 160. 0 (100) Total machine hours 220. 8 (69. 0) 35. 2 (11. 0) 30. 4 (9. 5) 33. 6 (10. 5) 320. 0 (100) $/hour 20 20 20 20 20 Total machine costs $4,416 (69. 0) 704 (11. 0) 608 (9. 5) 672 (10. 5) $6,400 (100) Quality cost measurement 745 Note: Figures in parentheses are labor hours indicated as a percentage Table VII. Assigning machine resource costs to activities hich provide the greatest opportunities for improvement. In view of this, the department manager requests quality engineers and industrial engineers to investigate the feasibility of changing 100 percent inspection to sampling inspection and to develop the methods of reducing set-up time. In addition, total cost of the PAF-related activities, inspection ($2,088), warranty repair ($1,120), rework ($1,104), and maintenance ($530), accounts for 27. 36 percent ($4,842/$17,696) of total manufacturing cost excluding direct material cost; the first three PAF-related activities are non-value-added and their cost accounts for 24. 37 percent ($4,312/$17,696).
This indicates that there are great opportunities for reducing quality costs and that there is an emerging need to identify where the opportunities lie, illustrated in the next subsection. Note that the company in this example separates idle capacity costs from activity cost calculation. This is not for external financial reporting and just for internal decision making. It will let managers know how much idle capacity costs, push managers to deploy the unused resources, and avoid distorting product costs. Activity cost assignment In the second stage of ABC cost assignment view, activity costs are traced to cost objects. ABC uses activity drivers to measure the consumption of activities by cost objects. In this example, products are used as the cost objects. This can trace COQ-related and COQ-unrelated costs to products.
For this example, the following data are shown in columns (5)-(10) of Table VIII: • • • the activity driver quantities of various activities consumed by Product A and Product B; the costs per activity driver of various activities; and the activity costs of various activities traced to Product A and Product B. Activities Machining Rework Warranty repair Inspection Package Scheduling Maintenance Set-up Material handling Idle Total Notes: 1. Formulas for calculation: (4) = (1) + (2) + (3); (7) = (5) + (6); (8) = (4)/(7); (9) = (5) ? (8); (10) = (6) ? (8); 2. The percentages within parentheses of columns (9) and (10) are the percentages of total product cost Table VIII. Assigning activity costs to products Labor (1) 2,000 400 512 1,392 1,536 208 448 1,600 672 832 9,600 672 6,400 82 20 294 Activity costs ($) Machine Others (2) (3) 4,416 704 608 696 304 Total (4) 6,416 1,104 1,120 2,088 1,840 208 530 1,920 966 1,504 17,696 Activity drivers Activity driver quantity Product A Product B Total (5) (6) (7) 124. 2 43 20 450 225 18 124. 2 6 9 – 96. 6 32 12 350 350 8 96. 6 2 14 220. 8 75 32 800 575 26 220. 8 8 23 Machine hours # of reworks # of warranty repairs # of tests # of units # of batches Machine hours # of set-ups # of moves – 29. 06 14. 72 35. 00 2. 61 3. 20 8. 0 2. 40 240. 00 42. 00 – 1,696 – – Total activity cost Direct material cost Total product cost Production quantity Unit product cost $/activity Activity cost assignment ($) driver Product A Product B (8) (9) (10) 3,609. 00 (32. 13) 632. 96 (5. 63) 700. 00 (6. 23) 1,174. 0 (10. 45) 720. 00 (6. 41) 144. 00 (1. 28) 298. 13 (2. 65) 1,440. 00 (12. 82) 378. 00 (3. 36) – 9,096. 59 (80. 97) 2,137. 50 (19. 03) 11,234. 09 (100) 225 49. 93 2,807. 00 (28. 88) 471. 04 (4. 85) 420. 00 (4. 32) 913. 50 (9. 40) 1,120. 00 (11. 52) 64. 00 (0. 66) 231. 87 (2. 39) 480. 00 (4. 94) 588. 00 (6. 05) – 7,095. 41 (72. 99) 2,625. 00 (27. 01) 9,270. 41 (100) 350 27. 77 746 IJQRM 15,7 This Table includes total activity cost, direct material cost, total product cost, and unit product cost for each product. By using the resulting data shown in columns (9) and (10), the department can prepare a product cost analysis report as shown in Table IX.
This report portrays product costs by direct material cost and activity costs, which is completely different from traditional cost report. This example report gives the department manager the following implications: • For the overall figures, total non-value-added cost accounts for 32. 05 percent of total manufacturing cost; total COQ accounts for 21. 56 percent. Thus, there will be great opportunities for improvement. Product A incurs more non-value-added cost/quality cost than product B. • For two high-cost activities, inspection and set-up, Product A incurs more cost than Product B, especially for set-up. Thus, reducing set-up time for Product A has the first priority in improving the set-up activity. Quality cost measurement 747
ABC Company Department: XYZ Product Cost Analysis Report Product A Unit produced Direct material cost ($) Activity costs ($) Machining (VA) a Package (VA) Maintenance (VA/COQ) Inspection (NVA/COQ) Rework (NVA/COQ) Warranty repair (NVA/COQ) Set-up (NVA) Material handling (NVA) Scheduling (Gray) Total activity cost Total product cost ($) Idle capacity cost Total manufacturing cost Unit product cost Total value-added cost Total non-valueadded cost Total COQ Total COQ per unit a Product B 350 2,625. 00 (27. 01) 2,807. 00 (28. 88) 1,120. 00 (11. 52) 231. 87 (2. 39) 913. 50 (9. 40) 471. 04 (4. 85) 420. 00 (4. 32) 480. 00 (4. 94) 588. 00 (6. 05) 64. 00 (0. 66) 7,095. 41 (72. 99) 9,720. 41 (100) 27. 77 4,158. 87 (42. 78) 2,872. 54 (29. 55) 2,036. 41 (20. 95) 5. 82 Total 225 2,137. 50 (19. 03) 3,609. 00 (32. 13) 720. 00 (6. 41) 298. 13 (2. 65) 1,174. 50 (10. 45) 632. 96 (5. 63) 700. 00 (6. 23) 1,440. 00 (12. 82) 378. 00 (3. 36) 144. 00 (1. 28) 9,096. 59 (80. 97) 11,234. 09 (100) 49. 3 4,627. 13 (41. 19) 4,325. 46 (38. 50) 2,805. 59 (24. 97) 12. 47 4,762. 50 (21. 21) 6,416. 00 (28. 57) 1,840. 00 (8. 19) 530. 00 (2. 36) 2,088. 00 (9. 30) 1,104. 00 (4. 92) 1,120. 00 (4. 99) 1,920. 00 (8. 55) 966. 00 (4. 30) 208. 00 (0. 93) 16,192. 00 (72. 10) 1,504. 00 (6. 70) 22,458. 50 (100) 8. 786. 00 (39. 12) 7,198. 00 (32. 05) 4,842. 00 (21. 56) Table IX. Product cost analysis report Notes: VA = value-added; NVA = non-value-added; COQ = cost of quality Figures in parentheses indicate a percentage IJQRM 15,7 • For the non-value-added and failure-cost-related activities, rework and warranty repair, Product A incurs more cost than Product B. 748
Conclusions While most COQ measurements methods are activity/process oriented, traditional cost accounting establishes cost accounts by the categories of expenses, instead of activities. Thus, many COQ elements should be estimated or collected by other methods. The main deficiencies of most COQ systems in measuring COQ are: • no consensus method to allocate overhead costs to COQ elements; • the failure to trace quality costs to their sources; and • the lack of information about how indirect workers spend their time on various activities. These deficiencies can be easily overcome under ABC together with work sampling. Based on the similarities of COQ approaches and ABC, this paper proposes an integrated COQ-ABC framework.
Ideally, ABC and COQ systems should be merged as one and share the common database in order to supply various cost and nonfinancial information for the related management techniques, ABM, TQM, and BPR. In addition, the integrated COQ-ABC information system should be integrated with the existing company accounting system eventually in order to reduce the resources required to manage the system (Goulden and Rawlins, 1995) and to avoid data redundancy and inconsistency. Under ABC, quality costs are achieved in the first stage of ABC cost assignment view and then traced to their sources in the second stage of ABC cost assignment view. ABC uses nested activity centers to create a multi-tier COQ reporting system to meet various management’s information needs and to support COQ responsibility accounting.
The cost and nonfinancial information achieved from the integrated COQ-ABC system can be used to identify the magnitude of the quality improvement opportunities, to identify where the quality improvement opportunities exist, and to continuously plan the quality improvement programs and control quality costs. The long-term goal of the integrated COQ-ABC system is to eliminate non-value-added activities, which are related to appraisal and failure costs for the PAF approach and CONC and some COC for the process cost approach, and to streamline value-added activities/processes. Moreover, the ultimate goal will be to continuously improve processes/activities/quality so that no defects at all are produced and quality cost measurement ultimately becomes unnecessary. References Albright, T. L. nd Roth, H. P. (1994), “Managing quality through the quality loss function”, Journal of Cost Management, Vol. 7 No. 4, pp. 20-8. ASQC Quality Costs Committee (1974), Quality costs – What and How, American Society for Quality Control, Milwaukee, WI. Beischel, M. E. (1990), “Improving production with process value analysis”, Journal of Accountancy, Vol. 170 No. 3, pp. 53-7. Bohan, G. P. and Horney, N. F. (1991), “Pinpointing the real cost of quality in a service company”, National Productivity Review, Vol. 10 No. 3, pp. 309-17. Brimson, J. A. (1991), Activity Accounting – An Activity-Based Costing Approach, John Wiley ; Sons, Inc. , New York, NY.
BS 6143: Part 2 (1990), Guide to Economics of Quality: Prevention, Appraisal and Failure Model, British Standards Institution, London. BS 6143: Part 1 (1992), Guide to the Economics of Quality: The Process Cost Model, British Standards Institution, London. Burgess, T. F. (1996), “Modelling quality-cost dynamics”, International Journal of Quality ; Reliability Management, Vol. 13 No. 3, pp. 8-26. Carr, L. P. (1992), “Applying cost of quality to a service business”, Sloan Management Review, Vol. 33 No. 4, pp. 72-7. Chen, Y. -S. and Tang, K. (1992), “A pictorial approach to poor-quality cost management”, IEEE Transactions on Engineering Management, Vol. 39 No. 2, pp. 149-57. Cooper, R. 1988), “The rise of activity-based costing – Part I: what is an activity-based cost system? ”, Journal of Cost Management, Vol. 2 No. 2, pp. 45-54. Cooper, R. (1990), “Cost classification in unit-based and activity-based manufacturing cost systems”, Journal of Cost Management, Vol. 4 No. 3, pp. 4-14. Cooper, R. and Kaplan, R. S. (1988), “Measure costs right: make the right decisions”, Harvard Business Review, Vol. 66 No. 5, pp. 96-103. Crosby, P. B. (1984), Quality Without Tears, Penguin Books, Ontario. Crossfield, R. T. and Dale, B. G. (1990), “Mapping quality assurance systems: a methodology”, Quality and Reliability Engineering International, Vol. 6 No. 3, pp. 167-78. Dale, B. G. and Plunkett, J. J. 1991), Quality Costing, Chapman ; Hall, London. Feigenbaum, A. V. (1956), “Total quality control”, Harvard Business Review, Vol. 34 No. 6, pp. 93-101. Goulden, C. and Rawlins, L. (1995), “A hybrid model for process quality costing”, International Journal of Quality ; Reliability Management, Vol. 12 No. 8, pp. 32-47. Hammer, L. H. , Carter, W. K. and Usry, M. F. (1993), Cost Accounting, 11th ed. , South-Western Publishing Co. , Cincinnati, Ohio. Hammer, M. and Champy, J. (1993), Reengineering the Corporation – A Manifesto for Business Revolution, Harper Business, New York, NY. Harrington, H. J. (1993), “Process breakthrough: business process improvement”, Journal of Cost Management, Vol. No. 3, pp. 30-43. Johnson, M. A. (1995), “The development of measures of the cost quality for an engineering unit”, International Journal of Quality ; Reliability Management, Vol. 12 No. 2, pp. 86-100. Juran, J. M. (1951), Quality Control Handbook, 1st ed. , McGraw-Hill, New York, NY. Juran, J. M. , Gryna, F. M. and Bingham, R. (1975), Quality Control Handbook, 3rd ed. , McGraw-Hill, New York, NY. Lawson, R. A. (1994), “Beyond ABC: process-based costing”, Journal of Cost Management, Vol. 8 No. 3, pp. 33-43. Lund, J. (1990), “Using EXCEL spreadsheet software to design and conduct a work”, Industrial Engineering, Vol. 22 No. 1, pp. 47-9. Marsh, J. 1989), “Process modelling for quality improvement”, Proceedings of the Second International Conference on Total Quality Management, IFS Publications, Bedford, pp. 111-21. Miller, J. A. (1992), “Designing and implementing a new cost management system”, Journal of Cost Management, Vol. 5 No. 4, pp. 41-53. Quality cost measurement 749 IJQRM 15,7 750 Miller, J. A. (1996), Implementing Activity-Based Management in Daily Operations, John Wiley ; Sons, New York, NY. Oakland, J. S. (1993), Total Quality Management, 2nd ed. , Butterworth-Heinemann Ltd, Oxford. O’Guin, M. C. (1991), The Complete Guide to Activity Based Costing, Prentice-Hall, Englewood Cliffs, NJ. Ostrenga, M. R. 1991), “Return on investment through the cost of quality”, Journal of Cost Management, Vol. 5 No. 2, pp. 37-44. Ostrenga, M. R. and Probst, F. R. (1992), “Process value analysis: the missing link in cost management”, Journal of Cost Management, Vol. 6 No. 3, pp. 4-13. Plunkett, J. J. and Dale, B. G. (1987), “A review of the literature on quality related costs”, International Journal of Quality ; Reliability Management, Vol. 4 No. 1, pp. 40-52. Plunkett, J. J. and Dale, B. G. (1988), “Quality costs: a critique of some ‘economic cost of quality’ models”, International Journal of Production Research, Vol. 26 No. 11, pp. 1713-26. Porter, L. J. and Rayner, P. 1992), “Quality costing for total quality management”, International Journal of Production Economics, Vol. 27, pp. 69-81. Raffish, N. and Turney, P. B. B. (1991), “Glossary of activity-based management”, Journal of Cost Management, Vol. 5 No. 3, pp. 53-63. Ravitz, L. (1991), “The cost of quality: a different approach to noninterest expense management”, Financial Managers’ Statement, Vol. 13 No. 2, pp. 8-13. Richardson, W. J. (1976), Cost Improvement, Work Sampling, and Short Interval Scheduling, Reston Publishing Company, Inc. , Reston, Virginia. Ross, D. T. (1977), “Structured analysis (SA): a language for communicating ideas”, IEEE Transactions on Software Engineering, Vol. SE-3 No. 1, pp. 16-34. Ross, J. E. and Wegman, D. E. 1990), “Quality management and the role of the accountant”, Industrial Management, Vol. 32 No. 4, pp. 21-3. Schneiderman, A. M. (1986), “Optimum quality costs and zero defects”, Quality Progress, November, pp. 28-31. Shank, J. K. and Govindarajan, V. (1994), “Measuring the ‘cost of quality’: a strategic cost management perspective”, Journal of Cost Management, Vol. 8 No. 2, pp. 5-17. Sharman, P. (1996), “Activity/process budgets: a tool for change management”, CMA Magazine, Vol. 70 No. 2, pp. 21-4. Simpson, J. B. and Muthler, D. L. (1987), “Quality costs: facilitating the quality initiative”, Journal of Cost Management, Vol. 1 No. 1, pp. 25-34. Son, Y. K. and Hsu, L. -F. 1991), “A method of measuring quality costs”, International Journal of Production Research, Vol. 29 No. 9, pp. 1785-94. Thorne, H. C. (1990), “The cost of quality”, 1990 AACE Transactions. Tsai, W. -H. (1994), “Product-mix decision model under activity-based costing”, Proceedings of 1994 Japan-U. S. A. Symposium on Flexible Automation, Vol. I, Institute of System, Control and Information Engineers, Kobe, Japan, 11-18 July, pp. 87-90. Tsai, W. -H. (1996a), “A technical note on using work sampling to estimate the effort on activities under activity-based costing”, International Journal of Production Economics, Vol. 43 No. 1, pp. 11-16. Tsai, W. -H. 1996b), “Activity-based costing model for joint products”, Computers ; Industrial Engineering, Vol. 31 No. 3/4, pp. 725-9. Turney, P. B. B. (1991), Common Cents: The ABC Performance Breakthrough – How to Succeed with Activity-Based Costing, Cost Technology, Hillsboro, Oregon. Turney, P. B. B. (1992), “What an activity-based cost model looks like”, Journal of Cost Management, Vol. 5 No. 4, pp. 54-60. Appendix: Glossary of ABC/ABM terms mentioned in this paper (Raffish and Turney, 1991) Activity. 1. Work performed within an organization. 2. An aggregation of actions performed within an organization that is useful for purposes of activity-based costing. Activity analysis. The identification and description of activities in an organization.
Activity analysis involves determining what activities are done within a department, how many people perform the activities, how much time they spend performing the activities, what resources are required to perform the activities, what operational data best reflect the performance of the activities, and what value the activity has for the organization. Activity analysis is accomplished by means of interviews, questionnaires, observation, and review of physical records of work. Activity attributes. Characteristics of individual activities. Attributes include cost drivers, cycle time, capacity, and performance measures. For example, a measure of the elapsed time required to complete an activity is an attribute. Activity cost pool. A grouping of all cost elements associated with an activity. Activity driver. A measure of the frequency and intensity of the demands placed on activities by cost objects. An activity driver is used to assign costs to cost objects.
It represents a line-item on the bill of activities for a product or customer. An example is the number of part numbers, which is used to measure the consumption of material-related activities by each product, material type, or component. The number of customer orders measures the consumption of order-entry activities by each customer. Sometimes an activity driver is used as an indicator of the output of an activity, such as the number of purchase orders prepared by the purchasing activity. Activity-based costing (ABC). A methodology that measures the cost and performance of activities, resources, and cost objects. Resources are assigned to activities, then activities are assigned to cost objects based on their use.
Activity-based costing recognizes the causal relationships of cost drivers to activities. Activity-based management (ABM). A discipline that focuses on the management of activities as the route to improving the value received by the customer and the profit achieved by providing this value. This discipline includes cost driver analysis, activity analysis, and performance measurement. Activity-based management draws on activity-based costing as its major source of information. Best practices. A methodology that identifies an activity as the benchmark by which a similar activity will be judged. This methodology is used to assist in identifying a process or technique that can increase the effectiveness or efficiency of an activity.
The source may be internal (e. g. taken from another part of the company) or external (e. g. taken from a competitor). Another term used is competitive benchmarking. Cost driver. Any factor that causes a change in the cost of an activity. For example, the quality of parts received by an activity (e. g. the percent that are defective) is a determining factor in the work required by that activity because the quality of parts received affects the resources required to perform the activity. An activity may have multiple cost drivers associated with it. Cost driver analysis. The examination, quantification, and explanation of the effects of cost drivers.
Management often uses the results of cost driver analyses in continuous improvement programs to help reduce throughput time, improve quality, and reduce cost. Cost element. An amount paid for a resource consumed by an activity and included in an activity cost pool. For example, power cost, engineering cost, and depreciation may be cost elements in the activity cost pool for a machine activity. Cost object. Any customer, product, service, contract, project, or other work unit for which a separate cost measurement is desired. Non-value-added activity. An activity that is considered not to contribute to customer value or to the organization’s needs. The designation “non-value-added” reflects a belief that the activity Quality cost measurement 751 IJQRM 15,7 752 an be redesigned, reduced, or eliminated without reducing the quantity, responsiveness, or quality of the output required by the customer or the organization. Pareto analysis. The identification and interpretation of significant factors using Pareto’s rule that 20 percent of a set of independent variables is responsible for 80 percent of the result. Pareto analysis can be used to identify cost drivers or activity drivers that are responsible for the majority of cost incurred by ranking the cost drivers in order of value. Performance measures. Indicators of the work performed and the results achieved in an activity, process, or organizational unit. Performance measure may be financial or nonfinancial.
An example of a performance measure of an activity is the number of defective parts per million. An example of a performance measure of an organizational unit is return on sales. Process. A series of activities that are linked to perform a specific objective. For example, the assembly of a television set or the paying of a bill or claim entails several linked activities. Resource. An economic element that is applied or used in the performance of activities. Salaries and materials, for example, are resources used in the performance of activities. Resource driver. A measure of the quantity of resources consumed by an activity. An example of a resource driver is the percentage of total square feet occupied by an activity.
This factor is used to allocate a portion of the cost of operating the facilities to the activity. Value-added activity. An activity that is judged to contribute to customer value or satisfy an organizational need. The attribute “value-added” reflects a belief that the activity cannot be eliminated without reducing the quantity, responsiveness, or quality of output required by a customer or organization. Value analysis. A cost-reduction and process-improvement tool that utilizes information collected about business processes and examines various attributes of the process (e. g. diversity, capacity, and complexity) to identify candidates for improvement efforts.