Evaluation System for Autonomous Control Methods in Coupled Planning and Control Systems

Abstract

Autonomous control is an approach to cope with increasing complexity in production and logistics. Autonomous control methods are characterised by decentralised coordination of logistic objects in a heterarchical organisation structure. Autonomous objects themselves are capable of processing information in order to make and execute decisions on their own. Up to now, research in this field mainly focuses on the effects of autonomous control methods disregarding their integration in existing planning systems. Thereby, it is currently not possible to give a substantiated recommendation, which combination of planning and autonomous control methods achieves a sufficient degree of logistic objective achievement in production systems of different complexity. Existing evaluation systems in the field of autonomous control methods remain on a mainly qualitative level disregarding the planning system. Therefore, this paper presents a quantitative, three-dimensional evaluation system. First, it operationalises the degree of complexity in production systems depending on shop floor conditions and disturbances. Second, the paper operationalises the degree of autonomy of production systems. Thereby, it considers the type and intensity of coupling between the planning and control level. Third, a vector of performance indicators is defined to measure the logistic objective achievement and the degree of plan fulfilment. The result is an evaluation system which allows a complete quantitative evaluation of autonomous control in production systems. Furthermore, the focus of consideration is expanded from the mere control level to the planning and control level. The influence of the strength of coupling between planning and control methods on the autonomy is also taken into account and operationalised. Finally, the evaluation system is exemplarily applied to a published simulation study.