Poor quality costing

Abstract: This thesis presents an effective model for measuring the cost of poor quality. It consists of two models, developed during the research project, assessments and the construction of measurement systems. By tradition, most companies prefer to design systems of their own for measuring their poor quality costs, but experience shows that this is difficult and time consuming, and, none the less, there is always the risk of failure. With this experience as a base, the model for a measurement system, put forward in the thesis, has been constructed and it should, consequently, have a better chance of success. The author would, however, like to point out that these measurement systems can only be used to measure a limited number of selected costs and should therefore only be applied when the company has matured some in its quality activities. Assessments, on the other hand, are simpler to perform, produce quick results, and, as shown in case studies, tend to be very successful. On account of that, the thesis highly recommends that assessments should be used as main means, particularly when the intention is to influence attitudes and behaviors, or to give priority to the most costly problems. The bulk of the research work has for this reason been focused on the development and test of a methodology for assessments, as past experiences and researches into these fields are very limited. Case studies show that the method developed for assessments has functioned quite well in the companies using it. Poor quality costs, studied in the area 9-16% of the business turnover, have been registered. This is far from the total poor quality costs of these businesses, but it is a considerably higher part than previously measured. The time spent on these assessments has been moderate, and the members of the assessment groups and the co-workers tended to respond in favor. The management groups, in the companies in question, have all taken an considerable interest in the studies and, consequently, in the results.