Vol.5 No.2 2012

26/66

Research paper : Toward the integrated optimization of steel plate production process (K. Nishioka et al.)−100−Synthesiology - English edition Vol.5 No.2 (2012) subjects matters covered by the planning.[12]Regarding the support systems of production control, MRP (material requirements planning) was introduced by General Electric in the mid-1950s, and since then for reinforcing the limitations of early MRP, support systems such as MRP , ERP (enterprise resource planning) and APS (advanced planning and scheduling) have been developed. Regarding their deployment and diffusion, the results of application of MRP and APS to simple and stabilized manufacturing processes have been reported. However, there is still no report on the result of their application to complex manufacturing processes where production volume changes frequently.[12]-[21] This reflects the fact that in the assembly industries where the pull-type production control represented by the lean production system is effective, the plan is reviewed in multi-phases and the accuracy of the plan of parts ordering, etc. are automatically improved as the time of actual production approaches.[12] Therefore, there has not been much need to build a comprehensive model/ support system that dynamically and organically connects plans with different time scales.In contrast, in the steel industry, a typical process industry, for operating integrated and continuous large-scale production facilities, an enormous amount of information for equipment control and production control has to be handled. Therefore, it introduced a production control system supported by large-scale computers ahead of other industries.[13] The production control system in the steel industry, corresponding fundamentally to the push-type production control system, prioritized the optimal control of the processes of intense heat and high temperatures, responses to the production fluctuation of upstream processes, and the maximization of each manufacturing lot size by integrating order information. Therefore, only very limited energy has been spent for supporting planning and scheduling for achieving the integrated optimization in terms of the manufacturing lead time and the in-process stock of intermediate goods. Furthermore, the sophistication and diversified specifications of recent products increased the complexity of the production control of plate manufacturing. Since large-scale production was carried out combining orders of various steel types whose processes to be completed were different, it was nearly impossible to determine at the very start of manufacturing the completion processes of individual intermediate products, and it was extremely difficult to predict and control manufacturing lead time.To realize the pull-type production control under such an environment, it is necessary to dynamically realize the optimization of manufacturing lot sizes and manufacturing lead time in all the processes by developing a model capable of defining comprehensively the influence of the size of manufacturing lots on the efficiency of each process and the manufacturing lead time, and by deploying support systems. In the steel industry where production fluctuation and variation are substantial, applying the existing MRP, APS, etc. was practically impossible, so we developed a series of models and introduced them to the shop floor operation one by one. 5.2 Development of a new set of production control systems (1) Efficiency modelFor improvement in the efficiency of rolling processes (for rolling processes, tonnage of slabs rolled per hour), the rolling processes that are the mainline in plate manufacturing were always the most important elements. Conventional evaluation implemented concerned mainly the reinforcement of individual pieces of equipment, but investment in the reinforcement in a short cycle means a large extra load in terms of management, and this is not easy to achieve. To overcome the issues of heavier load in rolling as a consequence of expanding TMCP technology application and the lower efficiency caused by diversified products, we began concentrating our efforts on enhancing the efficiency of the entire rolling processes arranged continually, directly, and in tandem. The efficiency of each process in rolling varies greatly depending on product specifications. Moreover, friction of the preceding and subsequent material processing (idle waiting time accrued due to the difference in the preceding and subsequent processing time) frequently occurs, changing sequentially the bottleneck processes of the materials being processed. The importance of bottleneck countermeasures is clearly indicated in the theory of constraints (TOC),[14] but because it is difficult to apply simple bottleneck countermeasures to rolling processes, we developed a new efficiency model that enables quantitative evaluation of the lowering of efficiency attributable to the friction between individual rolling processes. The development of this efficiency model was based on the assumption that the TOC is applicable also to the solution of problems in complex and continued processes like rolling processes.(2)Manufacturing lead time modelThanks to the efficiency model, the optimization of process design and appropriate equipment reinforcement of bottleneck processes were realized and the production efficiency of the rolling processes improved dramatically. This, however, verified the lack of capacity in the finishing processes, resulting in the increase of in-process stock caused by the fluctuation in the amount to be processed. To deal with this, we worked on the efficiency improvement of each process throughout the finishing processes, and with the help of a research and development crew, developed a logistics simulator using a simulation tool for modeling discrete event systems, in an attempt to reduce in-process stock and shorten the production lead time. However, we did not succeed

※このページを正しく表示するにはFlashPlayer9以上が必要です