Summary by Michele Martinez
Ph.D. Program in Accounting
University of South Florida, Spring 2002
In this study the authors’ report on a field-based study on the differences for a single division of a facility that utilizes just-in-time (JIT) manufacturing. In this study the authors’ divide the research project up into two parts. In the first part, the authors’ objective was to link the behavioral variables collected from direct laborers to performance measures assessed by the firm. The second stage of the study consisted of using the performance measures to explain the variation present in the overall measures of workgroup performance or workgroup ratings.
The first stage of the project developed a survey instrument that was administered to direct laborers to capture their perceptions of JIT and the company’s performance measures. This information was then used as the independent variables for the study and as a source of explanation for the six performance measures: cycle time, defective, yield rate, schedule adherence, process problems, and cost efficiency. In addition, in the second stage, all of the performance measures were used as independent variables to explain workgroup ratings. In analyzing the results, the authors’ found that they were not as expected. In an attempt to understand why this was true they discovered that a number of important factors and events had transpired during the course of their study and that in actuality the site was extremely different from how it had been initially portrayed to them. Consequently, this paper tries to explain cross-sectional differences but also presents a critical analysis of both the methodological approach the authors’ utilized and the types of compromises and challenges they faced while conducting the research.
Our approach to field-based research
In an attempt to conduct field-based research and study JIT, the authors’ decided to study a single organizational unit. This choice was made in order to minimize the confounding effects found in most across-firm, cross-sectional studies. This type of design controls for many external-confounding effects and specific to this case yields a large individual sample for data collection. Thus, this approach differentiated itself from other management accounting field studies in that it had the potential to interview many key informants.
When conducting field-based research site selection is the most important and critical decision. While, the findings may not be very generalizable, the richness of the field study offers substantial insight elsewhere. The site used in this study was suggested to the authors’ during a MBA course that one of the authors’ taught. Entry to this organization was gained after several meetings and presentations. Two studies were proposed to the organization, however, circumstances allowed for only the cross-sectional study of a large division just converted to JIT.
Nature of the research site
The site utilized in this study was a division of a firm in the electronics industry. The division performs the final assembly of all major products that the company manufacturers. The division has approximately 800 employees divided into 30 direct labor workgroups and 19 indirect workgroups, each with a first-level manager. Workgroups within the division were the units of analysis, however, it was not made clear that management explicitly considered it a need for every workgroup to have a different set of performance measures.
Data Gathering and Controlling Confounding Effects
In order to collect all the necessary data for statistical analysis, the researchers spent parts of 20 days spread over 6 months inside the division. Contrary to most management accounting field studies, the time spent was used to collect information for development of a survey instrument that would be meaningful to employees.
There are six types of information that the firm consistently measures and uses to assess performance of workgroups, although not all workgroups are measured on all dimensions of performance.
1. Cycle time efficiency
2. Yield rate
3. Defective rate
4. Production schedule adherence
5. Product cost efficiency
6. Number of manufacturing process problems
One of the first big problems encountered was that some workgroups actually shared performance measures. As a result the sample size was drastically reduced and some workgroups may have received performance measures where it was not appropriate. Additionally, several measurement errors were discovered which constitute serious problems not only for the study but also for management.
Worker and Workgroup ratings
In order to obtain a workgroup measure, direct labor ratings were averaged. However, these ratings were all based on social and non-quantified, subjective dimensions of performance. The data was not organized in a way that would allow each manager to use it to assess performance.
As previously mentioned one of the major benefits of a within organization field study is the controlling for confounding effects. However, in this organization the division was constantly revising processes during the time of development of the data-gathering instrument. Due to time already spent with the organization the authors’ hoped that the results would still be interesting and publishable. Through interviews the following independent variables were selected:
1. Incentives – to late into the study, management decided to end its practice of incentives, which caused a serious deficiency in the study.
2. Information – gaps
3. Manufacturing practices
4. Individual or workgroup characteristics
Development of the survey instrument
Most of the scales utilized in this study are unique to this study; only a few items were derived from items used in other studies. However, due to time constraints the authors were unable to perform a complete testing of the instrument.
There were a total of 387 direct laborers usable data collected. This pool represents all but 30 direct labor workgroups.
Performance regression results
All of the models are plagued with low degrees of freedom, caused by the firm’s aggregation of performance measures.
Considerable collinearity among some of the independent variables.
Cycle time efficiency cannot be explained by the model
Defectives performance could not be explained except with cost efficiency
Yield performance was explained with problem solving ability
Schedule performance was only explained with exogenous EC
Cost performance was not explained sufficiently
Process performance had the greatest statistical significance
Overall except for Ecs few other factors affect performance.
Overall, this paper tests the affects of information on selected measures of JIT performance at a single site. This paper presents the field research results from a single site of one division of a firm that was designed to provide evidence on factors that affect performance across direct labor workgroups.
1. Researcher’s contacts should assess whether any major structural changes are apparent.
2. Develop enough time to test instruments.
3. Researchers should attempt to attain clearance to conduct the research project from the senior most executive possible.
4. Researcher must determine if the site will produce a large enough sample size.
5. The data that one obtains may not tie directly to the research concept.
1. Managers may not be honest if they are scared that the researchers may uncover something
1. The researcher may completely misread the environment.
2. The researcher may put together a survey instrument that is completely not reflective of what firm employees are experiencing.
3. The limits of allowable inquiry also need to be established.
Borthick, A. F., P. L. Bowen, and M. C. Sullivan. 1998. Controlling JIT II: Making the system monitor itself. Journal of Cost Management (July/August): 33-41. (Summary).
Crusoe, J., G. Schmelzle, and T. E. Buttross. 1999. The hidden costs of adopting JIT manufacturing. Journal of Cost Management (December): 23-26. (Summary).
Davenport, T. H. and J. Glaser. 2002. Just-in-time delivery comes to knowledge management. Harvard Business Review (July): 107-111. (Summary).
Deluzio, M. C. 1993. Management accounting in a just-in-time environment. Journal of Cost Management (Winter): 6-15. (Summary).
Deluzio, M. C. 1993. The tools of just-in-time. Journal of Cost Management (Summer): 13-20. (Summary).
Foster, G. and C. T. Horngren. 1987. Cost accounting and cost management in a JIT environment. Management Accounting (June): 19-25. (Summary).
Fullerton, R. R. 2003. Performance measurement and reward systems in JIT and non-JIT firms. Cost Management (November/December): 40-47. (Summary).
Fullerton, R. R. and C. S. McWatters. 2002. The role of performance measures and incentive systems in relation to the degree of JIT implementation. Accounting, Organizations and Society 27(8): 711-735. (Summary).
Kalagnanam, S. S. and R. M. Lindsay. 1998. The use of organic models of control in JIT firms: Generalising Woodward's findings to modern manufacturing practices. Accounting, Organizations and Society 24(1): 1-30. (Summary).
Lee, J. Y. and J. K. Winch. 1998. From push to pull: Management's control system modification for manufacturing change. Advances in Management Accounting (6): 75-92. (Summary).
Lessner, J. 1989. Performance measurement in a just-in-time environment: Can traditional performance measurements still be used? Journal of Cost Management (Fall): 23-28. (Summary).
Martin, J. R. Not dated. Lean concepts and terms. Management And Accounting Web. https://maaw.info/LeanConceptsandTermsSummary.htm
Martin, J. R. Not dated. Profit Beyond Measure graphics and notes. Management And Accounting Web. https://maaw.info/ArticleSummaries/ArtSumJohnsonBromsGraphicsNotes.htm
Martin, J. R. Not dated. What is lean accounting? Management And Accounting Web. https://maaw.info/LeanAccounting.htm
McIlhattan, R. D. 1987. How cost management systems can support the JIT philosophy. Management Accounting (September): 20-26. (Summary).
O'Brien, J. and K. Sivaramakrishnan. 1994. Accounting for JIT: A cycle time-based approach. Journal of Cost Management (Fall): 63-70. (Summary).
Patell, J. M. 1987. Adapting a Cost accounting system to just-in-time manufacturing: The Hewlett-Packard Personal Office Computer Division. Accounting & Management Field Study Perspectives, edited by William J. Bruns, Jr. and R. S. Kaplan. Harvard Business School Press: 229-267. (Summary).
Swenson, D. W. and J. Cassidy. 1993. The effect of JIT on management accounting. Journal of Cost Management (Spring): 39-47. (Summary).
Vollmann, T. 1990. Changing manufacturing performance measurements. Proceedings of the Third Annual Management Accounting Symposium. Sarasota: American Accounting Association: 53-62. (Summary).