Activity-based costing and process modeling for cost-conscious product design: A case study in a manufacturing company

Katja Tornberg , Miikka Jmsen, Jari Paranko Industrial Management, Tampere University of Technology, P.O. Box 541, 33101 Tampere, Finland http://dx.doi.org/10.1016/S0925-5273(00)00179-1, How to Cite or Link Using DOI Permissions & Reprints

Abstract The most effective way to control costs is to design them out of the products. However, cost data is rarely available for product designers in a usable form. The aim of this study was to investigate the possibilities of activity-based costing and the modeling of design, purchasing and manufacturing processes in providing useful cost information for product designers. The hypothesis was that activity-based costing and process modeling might provide an effective tool for the evaluation of different design options. The study was conducted in a large Finnish manufacturing company. First, the most costly items of one

product's sub-assembly were studied in order to identify the activities needed to produce the items and to calculate their activity-based costs. Second, the processes, in other words the activity chains, were modeled with graphic flowcharts from product design, purchasing, and manufacturing departments. Finally, the applicability of activity-based cost information and process models to product designing practices was tested. The results of the study suggested that activity-based costing and process modeling provide a good starting point in heading toward more cost-conscious design. This way the designers learn the relationships between the activities performed in the organization and their associated costs. The development of a parametric cost estimation model based on activity-based costing and process modeling provides a challenge for future research. Keywords

Activity-based management; Cost management; Product design; Process modeling

1. Introduction

1.1. Product design and cost accounting The important role of product development in cost management is undeniable. However, the need for cost information in product development has not been sufficiently met. Problems include both the content as well as the form of information. examined the present state of cost accounting from a designer's point of view in four Finnish manufacturing companies. During the study both accounting professionals (information producers) and managers of designers (users of information) were interviewed. As a result 11 improvement opportunities were presented, of which two are presented here. First, cost information should be better combined with product structure and business process activities. Second, designers need reliable information on a product's total costs. It is common for designers to get information only on a product's direct costs. However, the decisions made on product design also affect indirect costs . Cost accounting should help product development to design products that carry the lowest possible total costs while also taking into consideration other important design criteria have argued that activity-based costing provides cost information on each of the company's production processes so that the costs of

different designs can be compared. According to them, activitybased costing (ABC) puts product costs on the backs of the engineers . Consequently, companies develop ABC systems to provide more relevant economic information for product designers . 1.2. Definitions A number of writers have provided definitions for a business process. According to Laakso a business process is a structured, measured set of activities and flows that use necessary resources of the organization to provide a specified output for a particular customer . Smart et al. [11] offer a general process classification framework. The framework is helpful in specifying the business processes as well as in determining the process boundaries. In this classification, the main process classes are the operating, support, and managing processes. The operating processes are defined as the cross-functional and value-adding chains having a direct customer interface. The four main operating processes are product development, sales and marketing, order fulfillment, and customer service. Furthermore, parts of the main operating processes can be divided into sub-processes. For example, a production process can be seen as a part of the order fulfillment

process. In addition to production, the order fulfillment process includes several other activities, such as customer inquiry, proposal, contract, assembly, delivery, installation and invoicing [10]. The objective of process modeling is to help companies to improve their business process understanding. A qualitative model that provides a graphical interpretation of the process captures the structure of a business process. The objective of qualitative process modeling is to visualize the process and to achieve a commonly agreed view of the process structure. Graphical process presentation facilitates communication between people and helps in developing a common conceptual model of what goes on in the studied process[10]. The graphical process models (Fig. 1) are formatted into horizontal bands to illustrate where an activity is performed and by whom. This way the involvement of various constituents of the overall process can be isolated. The process goes simultaneously from left to right and from the top down. The activities are presented with a rectangle and the branches of the process are presented with a diamond. A shadow under a rectangle indicates that an activity consists of several sub-processes. In this way the process model becomes a hierarchical entity.

Fig. 1. An example of a graphical process model. Figure options A quantitative analysis of quality, cost or time may be done after the process structure is visualized. The quantitative process modeling is a vital part of the in-depth process understanding. It identifies problem areas and helps to make improvement decisions based on more solid information [10]. In fact, quantitative process modeling may in our case be seen as equivalent to activity-based costing. This results from the fact that the activities of an ABC model and the process steps of a qualitative process model may be defined as identical. In summary, qualitative process modeling is the visual presentation of the process, while quantitative process modeling/ABC is the analysis of quality, cost or time of that process. It should be noticed that qualitative process modeling is not a precondition for activity-based costing. However, if this is done simultaneously with activity-based costing, important information

about the procedures of the organization can be obtained quite easily. 1.3. Aim of the study The aim of this study was to investigate the possibilities of activity-based costing and process modeling in providing useful cost information for product designers. The hypothesis was that activity-based costing and process modeling might provide an effective tool for the evaluation of different design options (Fig. 2).

Fig. 2. Research problem and thesis. Figure options The focus of the study is on cost accounting. The purpose was to examine the changing role of cost accounting in providing useful information for decision making by product designers. Though many of the current trends in product designing practices, such as concurrent engineering and design for manufacturing and assembly (DFMA), emphasize the role of costs, these methods are not studied here. Rather than searching directly for the improvement of product designing practices, the purpose was to

investigate the usefulness of activity-based cost information and information about the company's processes for product designers. The customer of this study is product design. As majority of the total product costs incurs from decisions made in product design, product designers should be seen as the key customers of cost information. This is why the usefulness of activity-based cost calculations and process models was tested on product designers. However, even though the causes for most of the product costs can be traced back to the decisions made in product design, the costs mainly emerge in other parts of the organization. Consequently, in order to receive as much information about the product's total costs as possible, activitybased costing and process modeling were carried out in addition to product design also in the purchasing and manufacturing functions of the company. 2. Methodology The study was carried out in a large Finnish manufacturing company. The company has a significant global position within its industry. An example product sub-assembly was chosen. First, the most costly items of the sub-assembly were studied in order to identify the activities needed to produce the items and to calculate their

activity-based costs. Second, the processes were modeled with graphic flowcharts from product design, purchasing, and manufacturing functions. Data for activity-based costing and process modeling were gathered from personal interviews. The company pointed out five product designers with whom to investigate the gathered and analyzed cost information. The designers were chosen for their extensive experience in their field. Their current tasks included e.g. product development, modification design, and modular design. Activity-based cost calculations and process models of two example items were presented to the designers. Detailed examples of this material are shown in the appendices. The designers had four days to explore the given material. At this point, a questionnaire with structured questions was administered. The questionnaire had two objectives: first to define the needs of the company's product designers regarding cost information and second to determine their opinions about the usefulness of activity-based cost calculations and process models. In order to obtain a better understanding of the matter, the questionnaires were supplemented with personal interviews. Interviews were conducted with all five product designers. However, under the time constraint, answers to the questionnaire

were received from only three product designers. In the personal interviews, special emphasis was given to the role of activitybased cost calculations and process models. In the activity-based cost calculations, the activity-based costs are divided into three main groups: the costs of life-cycle activities, the costs of batch activities, and the costs of volume activities (see Appendix B). Life-cycle activities are performed by product type. The costs of life-cycle activities can be allocated only to the product for which the activities are performed. Examples include programming the machines, drawing the item, and making the annual supplier contract. Batch activities are used by batches of products or any other cost objects. Costs of batch activities are allocated evenly to all the cost objects produced in the batch in question. Examples include setting up the machines, transferring the production lot, and checking the invoice. The costs of volume activities can be allocated directly to cost objects. The total costs of volume activities depend directly on the quantity of the cost object. Examples include machining and inspection of the goods received. A simplified example of an activity-based cost calculation is shown in Fig. 3.

Fig. 3. A simplified example of an activity-based cost calculation. Figure options The life-cycle of the example product is estimated to be five years and the annual production volume is 100 units, which makes a total of 500 units in 5 years. In batch activities the batch size varies according to the activity in question: for example, the setups are done for a lot size of 2 units, while checking the invoice is done for a lot size of 8 units. Volume activities are consumed by every unit produced. The activity cost is the cost of performing the activity once. The activity cost per unit is calculated by dividing the activity cost by units. For instance, the cost of programming the machines per unit is calculated by the following formula:

The numbers in the cost calculations are fictitious. Also, when comparing the amount of activity costs versus material costs, it should be noticed that in this study sales and marketing and administration functions, which typically make up most of the overheads, were ruled out. If these were also taken into account, the proportional amount of activity costs as compared with total costs would be much higher.

3. Results 3.1. Product designers needs The first part of the questionnaire focused on determining the product designers needs regarding cost information. The answers showed that the product designers cost accounting knowledge varied considerably. For example, when the procurement costs of an item were discussed, the only cost that was mentioned was the price paid for the item. However, this is only part of the total procurement cost of an item. Product designers make their daily designing decisions based primarily on a product's technical features. From the designer's point of view the most important criteria for products are top quality, durability, performance, and correspondence with the customer's specifications. Only one interviewee mentioned the final costs of the product as a criterion for the designing decision. On the other hand, the designers thought that their supervisors evaluate them on the basis of timely delivery of the design and potential warranty costs. The questionnaire, however, indicated that product designers generally think cost information is important for them. On the other hand, the designers were doubtful about whether their decisions could have any impact on indirect costs. Consequently,

they felt the cost information of manufacturing and product development functions to be most important, while the costs of purchasing, sales and marketing, and administration functions were seen as less significant or too difficult to estimate and include in their decision processes. The cost information currently available was seen to have many deficiencies. The problems and needs found in this study were very similar to those presented by Uusi-Rauva and Paranko [2]. The only problems identified in their study that did not come out during the interviews were support in make or buy decisions and support in the search for raw materials or suppliers. On the other hand, this study revealed something that did not appear in their study: the need for feature-based cost information. Product designers were keen to know how the costs of a product were affected when, for example, tolerance was altered. The cost comparison between different production methods such as welding and machining was also seen as important. 3.2. The role of process models and activity-based cost calculations The latter part of the questionnaire was aimed at finding out the product designers opinions about the usefulness of the process models and activity-based cost calculations. As one of the major

benefits of the cost calculations the interviewees rated the fact that cost information was presented by product structure and by activities performed, which is usually not the case with traditional cost accounting. With the help of process models and cost calculations it was possible to estimate the effects of different design options on product costs. In addition to these cost-related advantages, there was still one which concerned specifically the process models: their role was seen as getting the whole picture of what goes on in the company. The models reflect the actual phases that the product goes through in the organization as it is being designed and produced. This led to an increased understanding of the activities needed to deliver the sub-assembly in question. The designers also pointed out the limitations of the process models and activity-based cost calculations. First, the information gathered was focused only on one sub-assembly and a limited amount of activities (product design, purchasing, and manufacturing). Thus, the scope of the study was quite narrow and did not provide a sufficient basis for evaluating the cost of other product components or sub-assemblies. Second, at present the cost calculations include too many activities. This makes the calculations too complicated and too

slow to use. However, it was considered extremely important to get more detailed information about the activities when needed. Third, the problem of the right level of information was seen to apply also to process models: the first picture should be quite simple, but it should also be possible to get more detailed information about design, purchasing, and manufacturing processes when needed. The designers provided suggestions on how to deal with these problems. Regarding the narrowness of the scope of the study, the solution suggested was to present the cost information by product parameters. With respect to the slowness and complexity of the cost calculations and the process models, the designers had two improvement suggestions: first, it would make sense to categorize the activities into larger groups. For example, all the life-cycle activities used by an item could be presented as only three activities: the life-cycle activities in manufacturing, the lifecycle activities in product design, and the life-cycle activities in purchasing. Second, it could be useful to combine the information on the cost calculations with the process models. An activity may be easy or more difficult to perform due to certain circumstances. Therefore, it would make sense that each step of the process model be given three different activity options: easy, ordinary, or difficult, each one with its own cost. Also the reasons which cause

the same activity to be easy or more difficult should be attached to the process model. In summary, the cost estimation model for product designers use should be simple at first sight, but it should also enable more detailed information to be obtained when needed. The model should be easy to use and provide information on-line. 4. Conclusions The results of this study lend support to the results of Uusi-Rauva and Paranko [2] regarding the need in product development for cost information. The differences which emerged between these two studies concerning needs result mainly from the different interviewee groups: the previous study concentrated on the needs of managers of designers (and accounting professionals), while in this study the focus was on the needs of product designers themselves. Consequently, the needs found in this study accentuated the daily work of the designer, that is, for example, cost information by product structure, by activities and also by product parameters. On the basis of this study, it can be argued that process modeling and activity-based costing can provide useful cost information for product designers. This derives basically from the following facts. First, the cost information is presented by product structure, which

is the form in which designers normally receive other important information in their daily work. Second, the process models and cost calculations reveal the effects of different design options on the whole order-to-delivery process, i.e. the total product costs. The calculations and process models created so far, however, are too complicated and too slow for daily use. The categorization of activities into larger groups and the attachment of cost information to the graphic process models provide a basis for solving this problem. Furthermore, on the basis of this information it is possible to estimate only the costs of a particular sub-assembly. Thus, feature-based costing, parametric cost estimation, etc., provide a challenge for future research. The critical role of product design in cost management is being recognized. Product design should be considered as one of the clients for a company's cost accounting; perhaps as one of the most important clients of this function. This demands a new perspective toward cost information: the content and form should vary according to who is the target of the information. Activitybased costing and process modeling provide one possible solution for providing product designers with the kind of cost information they desire and need. Appendix A. Process model

Figure options Appendix B. Calculation of activity-based costs Example item Units Activity cost Cost per unit % of total costs

Life-cycle activities Checking the drawing Designing the fastener Designing the tools Drawing the item 500 500 500 500 1000 2500 500 3500 2 5 1 7

Making the annual supplier contract Production routing Programming the machines Reviewing the annual supplier contract Selecting the suppliers Testing the program

500

2500

500 500

500 1000

1 2

500

1000

500 500

1500 500

3 1 29 2.5

Batch activities Checking the invoice Closing the work order Loading flexible manufacturing system Material handling Ordering the materials Receiving the materials Setting up the machine 8 2 2 2 2 11 0.25 1 5.5

2 8 2 2

30 23 12 29

15 2.875 6 14.5

45.125 Volume activities Inspection of delivery Machining with a lathe 67 252 67 252 319 Materials Activity costs Total costs 767 393.125 1160.125

3.9

27.5 66.1 33.9 100

Table options References 1.

o o

[1] E. Uusi-Rauva, J. Paranko, Cost accounting and needs of product development, Report No. 1/1998/Industrial Management, Tampere University of Technology, 1998 (in Finnish).

2.
o

[2]

E. Uusi-Rauva, J. Paranko, The role of cost accounting in engineering, Proceedings of the International Conference on Engineering Design, Tampere, Finland, 1997.

3.
o o

[3] R. Cooper, P. Turney, Internally focused activity-based cost systems. performance excellence in manufacturing and service organizations, Proceedings of the Third Annual Management Accounting Symposium, San Diego, CA, 1990.

4.
o o

[4] R.S. Kaplan, In defence of activity-based cost management, Management Accounting (1992) 5863.