Production Planning and Control with Learning Technologies: Simulation and Optimization of Complex Production Processes

Publisher Summary Since the early 1970s, order processing has increasingly been supported by engineering data processing (EDP) systems in German-speaking countries. This tremendously fast development in data processing has created the ability to cover the entire order processing chain, from customer request to delivery (invoicing). This chapter is based on order management, a new and general approach to production planning and control (PPC) in a turbulent environment. Its basis is the tight link between planning and execution activities in the order management cycle, and it includes an additional learning step. Because of a more reliable manufacturing forecast, more realistic production plans and therefore increased security in planning would follow. This is particularly important when establishing economically successful production in a turbulent environment. This chapter deals with the rough planning components at the individual site of a production network based on neural networks. In contrast to the classical planning procedure that is based on a constant lead-time, artificial neural networks (ANNs) allow planning with variable lead times, depending on the given situation. This chapter shows all of the requirements for such an approach to planning and its limits. The EDP system developed in the course of these tasks is drafted as an “add-on” to the company's existing PPC systems and is realized as a prototype.