Vol.5 No.2 2012

39/66

Research paper : Toward the integrated optimization of steel plate production process (K. Nishioka et al.)−113−Synthesiology - English edition Vol.5 No.2 (2012) processes, as the number of plates processed per hour.The rolling processes are comprised of the processes that are arranged continuously, in tandem and in direct connection and the processes are the slab yard process, heating process, roughing process, finish rolling process, accelerated cooling process and shearing process. In the slab yard process, slabs received from steelmaking are cut; in the heating process, the slabs are reheated; in the roughing process, the reheated slabs are rolled to the specified width; in the finish rolling process, the slabs rolled to the desired width are rolled to the specified thickness and length and the material property is integrated by controlled rolling; in the accelerated cooling process, the slabs after finish rolling are quenched using a large amount of cooling water, to integrate a quenched structure; and in the shearing process, the slabs after rolling/ cooling are divided. Steel plate products come in a wide variety of thicknesses, widths and lengths and of specifications in standards, and because the process conditions of each process are inevitably varied, the processing efficiency of each process varies significantly in response to the product specification. In other words, because the rolling processes are in a large scale and are also of mixed-flow production of many product types, and because the buffer between processes is small, friction between the preceding and subsequent material processing frequently occurs. Therefore, as the bottleneck processes found during processing change sequentially by every material, the efficiency of the overall rolling processes changes substantially, and therefore, it has been technically difficult to predict the efficiency accurately and easily. Most of the existing planning of production, manufacturing and processing used to be implemented by making various assumptions of conditions and calculating sequentially individual events that respond to their respective conditions, evaluating and comparing the results using the evaluation function and selecting the best fitting conditions. However, for large-scale and mixed-flow production of many product types, if the production permutation, i.e., the combination of simulation calculation conditions, becomes enormous, the load on the computer becomes excessive, and such a planning method is impractical.Each rolling process has different factors that work against good processing efficiency. For example, in the slab yard process, they are slab cutting speed, slab weight, etc.; in the heating process, charging temperatures of the furnace, heating conditions (temperature to extract slabs from the furnace, time to keep slabs), etc.; in the roughing mill process, the extracting temperature from the furnace, slab sizes, etc.; in the finish rolling process, rolling speed, waiting time between completions for controlling rolled structure, length of rolling, etc.; and in the shearing process, cutting speed, cutting accuracy, etc.In this efficiency model, these parameters that affect the processing efficiency of each process were extracted, and products were classified into groups for which these parameters have significant difference, and the processing efficiency was statistically calculated for every product group. In addition, a method was adopted to obtain the integrated efficiency specific to each product group by comparing the processing efficiency of each process for every product group classified in accordance with their product type, size, furnace charging temperature, etc., bottleneck processes in more than one sub-process arranged in tandem, in direct connection, and in multi-steps that are identified for each product group. 6 Integrated lead time modelQuestion(Motoyuki Akamatsu)It is stated that the enhancement of the rolling process efficiency increased the requirement fluctuation of the finishing processes, but is it correct in understanding that the enhancement of efficiency led to larger lot sizes and with the increase in the lot sizes, because the finishing processes are varied, such lot has to wait for processing in the finishing processes, thereby increasing in-stock volume?Regarding the variance of the spare delivery days, you state that the spare days for starting rolling against delivery date and the manufacturing lead time are independent, and the variance can be obtained by the sum of the respective variances. However, rolling start manufacturing completion is contained in the rolling start delivery date and it seems both are mutually dependent. Please additionally state why the spare days for starting rolling against the delivery date and the manufacturing lead time can be considered independent.Answer (Yasushi Mizutani)If the daily rolling volume increases thanks to the enhancement of the efficiency of the rolling process, inevitably the flow into the finishing processes, i.e., the required processing volume of the finishing processes, increases resulting in the increase of the fluctuation requirement. Your statement that “the enhancement of efficiency led to larger lot sizes” is correct. Larger variation of processing requirement in the finishing processes leads to longer waiting time and higher stock level.The production of steel plates is characterized, as already stated in the paper, by receiving orders for various products (products of the same specification in one order are about 3 t), manufacturing them in lots (condition of the same steelmaking: min. 300 t, but considering productivity, desirably over 2,000 t) and delivering products in the unit of a lot (for the same customer, delivery date, transportation means). The delivery date is specified for the same shipment lot, but as requested by the customer, orders of various products are contained in one shipment lot, and the manufacturing lot does not generally agree with the shipment lot. Therefore, even though the timing to start rolling is determined by counting backward the manufacturing lead time set from the delivery date, variation is inevitable as a result of grouping products in a manufacturing lot, and at the same time, variation associated with the fluctuation of operation in the upstream process of steelmaking that is independent from plates exists.In contrast, the manufacturing lead time is dependent on any processing requirement and on the operation fluctuation, and therefore, it is also with variation. Therefore, the spare delivery days are obtained as the difference in the spare days for starting rolling against delivery date (rolling start until delivery day) and the manufacturing lead time.Spare delivery days = spare days for starting rolling against delivery date (rolling start until delivery day) – manufacturing lead timeIf the two terms on the right are independent, the formula given in this paper of spare days until delivery = (spare days for starting rolling against delivery2 + manufacturing lead time2)1/2is valid. We verified this against past records to confirm that the relation expressed by the above formula is valid, and we concluded that the assumption of “the two terms on the right are independent” is quasi valid.

※このページを正しく表示するにはFlashPlayer9以上が必要です