Fuzzy logic control for mobile service systems with limited resources

It is common to control systems or plants through changing the values of their inputs in continuous or small discrete steps to achieve gradual change of the performance of the plant under control. However, when changing the inputs can only be achieved in large discrete values, problems arise on achieving stability of the plant under control as such large discrete changes in inputs cause large overshoots and oscillation at the outputs. More difficulties arise when multiple plants are competing simultaneously for limited resources of such large discrete values. This paper describes an approach for controlling such plants. The case study presented is to control a proposed system of mobile service units (MSUs) to be used in a mega-event, the Hajj, where different locations/zones will be competing simultaneously for a limited number of MSUs in order to provide adequate services for their continuously changing number of visitors. These MSUs are of high discrete values such that their provision or removal generates a significant change in the performance of the system. Fuzzy logic controllers (FLCs), representing each zone, are developed for issuing requests for MSUs based on their local situations. These requests are evaluated by the control centre using other sets of FLCs to consider global situations. An algorithm is used to respond to these evaluated requests according to their relative importance and issues orders to MSUs to move to the right zone at the right time to provide the appropriate amount of service for the existing number of visitors or what can be called 'just-in-time MSUs' (JIT-MSUs). Simulation results show that such a system can be controlled using the approach suggested by this paper and the response of the system varies with the values of its discrete inputs. Some conclusions and recommendations are drawn for the appropriate discrete input values and their control settings. The procedure for evaluating the different simulation runs for the different control settings is also presented.