Industry is generally keen to identify avenu.:s for phased investments in Information Technology (IT) in manufacturing enterprises. IT is a key enabler of extended enterprises involving oomputerization of the integrated manufacturing and logistics systems. Similar to the earlier efforts in CIM, the ultimate aim is to develop a decision and information system in organizations that provides an on -line control which is real time. A good example is the control of FMS (a mini-CIM) where we ensure an on -line control that is essentially real time control. Such capabilities are very expensive and industry is more interested in the phased development of such capabilities requiring phased investment. In our view Semi-Computerized Flexible Manufacturing (SCFM) systems (Wadhwa and Bhagwat. 1998] are a useful building block for the purpose. On-line control in SCFM systems involves decision and information delays where decisions are based on information which may be older than the real time information. To effectively control SCFM systems with defined levels of flexibility, it is essential to explicitly model and analyze the effect that decision and information delays may have on the performamce of a given on-line control strategy. In this respect cornputer simulation is an expedient approach. However this may be expensive in which concerns time, effort and costs as the number of factors in SCFM control and their possible levels may be too large. This may result in a very large number of possible combinations to simulate in order to identify optimal control directions. The controllers ideally require tools and methodologies that help them to quickly and effectively identify the priority factors and the impact of their interactions. In this paper we present one such approach to study the makespan perfomance of an SCFM system under a review period monitoring policy that entails variable information delays. The approach uses Taguchi methods to provide an expedient experimental platform for quick insights into the behavior of system under alternative factor level combinations. It thus allows the controllers to effectively and efficiently identify key factors and interactions to be controlled as a priority based on their relative coutributions to the makespan perfomance in a sample SCFM system. The results indicate that the relative contribution of decision delays on the makespan performance is maximum. Further the relative contribution of other factors such as routing flexibility and control strategy decreases as the review period increases. The implication for an SCFM controller is that the de ision and information delays must be well contained with a greater priority on the reduction in decision delays. We also highlight how the performance of the SCFM system may differ with that of the FMS. An important implication for the controllers is that the control knowledge of managing the FMS may not be a good substitute while managing SCFM systems. The efficacy of the simulation approach based on the Taguchi method is further proved when it is successfully able to capture the interaction effects between the key control factors and routing flexibility in the SCFM studied.
Review period, Decision delays, Routing flexibility, Control strategy, Factor Interactions, Taguchi methods, Simulation
All articles published by Studies in Informatics and Control (SIC) are published in accordance with the terms and conditions of the Creative Commons Attribution License (CC BY-NC-SA 4.0).