DEVELOPMENT OF SELF-MAINTENANCE PHOTOCOPIERS

The traditional reliability design methods are imperfect because the d esigned systems aim at fewer faults, but once a fault happens, the sys tems might hard fail. To solve this problem, we present a self-mainten ance machine (SMM), one that can maintain its functions flexibly even though faults occur. To achieve the capabilities of diagnosing and rep air planning, a model-based approach that uses qualitative physics was proposed. Regarding the repair-executing capability, a control-type r epair strategy was followed. A prototype of the SMM was developed, and it succeeded in maintaining its functions if the Structure did not ch ange. However, the prototype revealed the following problems when its reasoning system was used with a commercial product as embedded softwa re: (1) poor performance of the reasoning system, (2) system size that was too large, (3) low adaptability to environmental changes, and (4) roughness of qualitative repair operations. To solve these problems, we proposed a new reasoning method based on virtual cases and fuzzy qu alitative values. This methodology is one of knowledge compilation, wh ich gives better reasoning performance and can deal with real-world ap plications such as the SMM. By using this method, we finally developed a commercial photocopier that has self-maintainability and is more ro bust against faults. The commercial version has been supplied worldwid e as a product of Mita Industrial Co., Ltd., since April 1994.