Scholarlink

<p>Production planning and scheduling are critical for maintaining supply chain stability, particularly in flow shop production systems, which are structured so that products move sequentially through multiple stages of processing, commonly seen in manufacturing. While extensive research has been conducted on these systems, much of it relies on idealized assumptions that do not reflect real-world complexities. This study proposes an integrated model for optimizing production planning and distribution in a three-stage hybrid flow shop system, considering practical constraints such as batch scheduling, delivery time windows, warehouse and production machinery capacity limitations, vehicle and pallet availability, and setup times. The objective is to minimize costs related to production, setup, warehousing, transportation, and lost sales, thereby maximizing overall system profit. To solve this NP-hard problem, dynamic programming is employed alongside seven metaheuristic algorithms, including genetic algorithms, simulated annealing, harmony search, and four hybrid algorithms (SAGA, HSGA, HSGA-I, and HSGA-II). The performance of these algorithms is statistically analyzed and compared to dynamic programming on a small scale, demonstrating their effectiveness. The model and solution methods are further validated using a large-scale real-world case from the automotive industry supply chain, with results showing that HSGA-I outperforms other algorithms and offers practical applications for similar industrial challenges.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Production and Distribution Integrated Model for Batch Scheduling in a Hybrid Flow Shop Using Dynamic Programming and Metaheuristics

  • 期刊论文

  • 发表日期：

    2026年7月12日

查看全文
  • Kamran Dashti Maljaei,
  • S. Kamal Chaharsooghi,
  • Ali Husseinzadeh Kashan

摘要

Production planning and scheduling are critical for maintaining supply chain stability, particularly in flow shop production systems, which are structured so that products move sequentially through multiple stages of processing, commonly seen in manufacturing. While extensive research has been conducted on these systems, much of it relies on idealized assumptions that do not reflect real-world complexities. This study proposes an integrated model for optimizing production planning and distribution in a three-stage hybrid flow shop system, considering practical constraints such as batch scheduling, delivery time windows, warehouse and production machinery capacity limitations, vehicle and pallet availability, and setup times. The objective is to minimize costs related to production, setup, warehousing, transportation, and lost sales, thereby maximizing overall system profit. To solve this NP-hard problem, dynamic programming is employed alongside seven metaheuristic algorithms, including genetic algorithms, simulated annealing, harmony search, and four hybrid algorithms (SAGA, HSGA, HSGA-I, and HSGA-II). The performance of these algorithms is statistically analyzed and compared to dynamic programming on a small scale, demonstrating their effectiveness. The model and solution methods are further validated using a large-scale real-world case from the automotive industry supply chain, with results showing that HSGA-I outperforms other algorithms and offers practical applications for similar industrial challenges.