有效实施Seru生产前提下多能工培训需求最少的优化决策

Seru production is recognized a new production pattern to cope with a volatile market. Multi-skilled worker is the pre-condition of seru production, however, all of previous researches are based on the assumption of fully multi-skilled workers. The assumption means that it will take a large amount of time and cost to train workers, so that companies can not implement seru production within a short time. Therefore, companies wish to improve production efficiency by seru production under the minimal training demand of multi-skilled workers . The project aims to investigate the optimal methods for multi-skilled workers training with minimizing training demands under effectively implementing seru production. The findings of this project will not only make theoretical contribution to seru production, but also proposes the management insights on successful seru production for the companies which want to fast adopt seru system under partial multi-skilled workers. The achievement of the project will help domestic enterprises to increase the productivity. To this end, the research activities will be conducted in the following parts: (P1) Optimal method for the conversion from assembly line to divisional seru system with minimizing multi-skilled workers training demand and maximizing productivity of seru system; (P2) Optimal method for divisional seru system improvement with minimizing multi-skilled workers training demand and maximizing productivity improvement of seru system; (P3) Optimal method for divisional seru system with minimizing multi-skilled workers training demand and minimizing the productivity difference with the seru system with fully multi-skilled workers; (P4) Optimal training methods for multi-skilled worker training in seru production；and (P5) Measurement methods of training effect, training attitude and rational mechanism to manage trainees.

Seru(日本式单元)生产是一种新的生产模式，多能工是其必要条件，但目前的研究都基于全能工，而全能工需要大量的培训时间和成本，导致企业难以快速的实施Seru生产。因此，企业希望通过最少的多能工培训实现Seru生产并能有效的提高生产效率。本项目以Seru生产最少多能工培训需求的优化决策为研究对象，以能在最短培训时间内利用Seru生产最大化提高生产效率为研究目标。首先探索生产线向分割式Seru系统转化时系统效率最高且多能工培训需求最少的方法。其次探索现有分割式Seru系统改进时系统效率提升最高且多能工培训需求最少的方法。然后探索近似全能工Seru系统性能的非全能工分割式Seru系统多能工培训需求最少的方法。最后探索适合Seru生产的多能工培训方法，以及度量培训效果和分析培训态度的方法。我国正处在经济转型期，仍以流水线生产模式为主，若能成功实施Seru生产，将能提高生产效率，获得良好的经济效益。

以Seru生产最少多能工培训需求的优化决策为研究对象，从理论角度探索Seru生产中最少的多能工培训需求，但能达到Seru生产效率最大化的方法，以实现企业在最短时间内通过Seru系统提高生产效率。取得了比较丰富的研究成果，研究成果概括如下：研究了分割式赛汝系统效率最高且多能工培训需求最少的优化方法，建立了分割式赛汝的makespan最小模型，开发了基于分解和减少解空间的精确算法（减少了99%的解），该精确算法分析了分割式赛汝系统性能和多能工需求的关系；研究了混合式赛汝系统效率最高且多能工培训需求最少的优化方法，开发了基于减小解空间和回溯的精确算法求解混合赛汝系统makespan最小下多能工最小的需求，用该精确算法分析各种情形下makespan的性能和多能工培训的需求；研究了总拖期最小的赛汝生产最优运作的方法，开发了精确算法和基于联合优化的智能算法；考虑到平衡性是赛汝系统的关键指标，研究了赛汝系统平衡性最优的赛汝生产运作方法；减少工人数能有效降低运营成本，研究了工人数最少且makespan不增加下的赛汝生产最优运作方法；开发了求解makespan最小赛汝生产系统最优运作的联合优化方法；鉴于调度规则对赛汝生产系统具有重要影响，研究了十种调度规则下赛汝系统最优运作解空间的复杂度，以及对赛汝生产系统运作结果的影响。