LEARNING BY OBJECTIVES FOR ADAPTIVE SHOP-FLOOR SCHEDULING

Effective production scheduling requires consideration of the dynamics and unpredictability of the manufacturing environment. An automated l earning scheme, utilizing genetic search, is proposed for adaptive con trol in typical decentralized factory-floor decision making. A high-le vel knowledge representation for modeling production environments is d eveloped, with facilities for genetic learning within this scheme. A m ultiagent framework is used, with individual agents being responsible for the dispatch decision making at different workstations. Learning i s with respect to stated objectives, and given the diversity of schedu ling goals, the efficacy of the designed learning scheme is judged thr ough its response under different objectives. The behavior of the gene tic learning scheme is analyzed and simulation studies help compare ho w learning under different objectives impacts certain aggregate measur es of system performance.