Owing to its high efficiency and flexibility, the seru production system (SPS), which originated in Japan, has attracted greater attention in management and academic studies. This research focuses on optimizing the configuration for implementing the SPS under an uncertain demand. The study is aimed at formulating a robust production system capable of effectively responding to stochastic demands. The primary issues are determining the amount of skill training required and matching workers with their corresponding skills. A stochastic optimization model is developed to minimize the total expected cost of the system, while considering the costs associated with training, staff shortage, and staff surplus. A heuristic algorithm is developed to solve this problem. Experimental results indicate that, compared to the full-skilled training strategy, appropriate partial skill training (such as the long chain skill training strategy) can yield greater benefits. The total cost and amount of skill training increase with growing differences in the product mix compositions, demand fluctuations, and number of product types. Moreover, the skill level of workers increases with a decrease in training cost and an increase in staff shortage and surplus costs.

中文翻译：

---

不确定需求下血清生产系统的优化技能配置

起源于日本的血清生产系统（SPS）由于其高效性和灵活性，在管理和学术研究中引起了更大的关注。本研究侧重于优化配置以在不确定需求下实施 SPS。该研究旨在制定一个能够有效应对随机需求的强大生产系统。主要问题是确定所需的技能培训量以及将工人与其相应的技能相匹配。开发随机优化模型以最小化系统的总预期成本，同时考虑与培训、人员短缺和人员过剩相关的成本。开发了一种启发式算法来解决这个问题。实验结果表明，与全技能训练策略相比，适当的部分技能培训（如长链技能培训策略）可以产生更大的收益。随着产品组合构成、需求波动和产品类型数量的差异越来越大，技能培训的总成本和数量也会增加。此外，工人的技能水平随着培训成本的降低以及人员短缺和过剩成本的增加而提高。