[论文解读] Efficient Data-Driven Production Scheduling in Pharmaceutical Manufacturing

Ioannis Balatsos, Athanasios Liakos+4 more|arXiv (Cornell University)|Feb 14, 2026

Scheduling and Optimization Algorithms被引用 0

一句话总结

该论文提出一个基于数据驱动、基于约束的优化框架，用于具有固定工艺路线、资源日历和清洗时间的药品工厂车间，在三个工业实例中相对于简化的参考方案实现了显著的总加工时间和延期时间降低。

ABSTRACT

This paper develops a data-driven, constraint-based optimization framework for a complex industrial job shop scheduling problem variant in pharmaceutical manufacturing. The formulation captures fixed routings and designated machines, explicit resource calendars with weekends and planned maintenance, and campaign sequencing through sequence-dependent cleaning times derived from site tables. The model is implemented with an open source constraint solver and evaluated on deterministic snapshots from a solid oral dosage facility under three objective formulations: makespan, makespan plus total tardiness, and makespan plus average tardiness. On three industrial instances of increasing size (10, 30, and 84 jobs) the proposed schedules dominate reference plans that solve a simplified variant without the added site rules. Makespan reductions reach \(88.1\%\), \(77.6\%\), and \(54.9\%\) and total tardiness reductions reach \(72.1\%\), \(58.7\%\), and \(18.2\%\), respectively. The composite objectives further decrease late job counts with negligible makespan change on the smaller instances and a modest increase on the largest instance. Optimality is proven on the small case, with relative gaps of \(0.77\%\) and \(14.92\%\) on the medium and large cases under a fixed time limit. The results show that a compact constraint programming formulation can deliver feasible, transparent schedules that respect site rules while improving adherence to due dates on real industrial data.

研究动机与目标

• Develop a data-driven, constraint-based optimization model for a complex pharmaceutical job shop with fixed routings and designated machines.
• Incorporate explicit resource calendars, weekends, planned maintenance, and sequence-dependent cleaning times into the scheduling model.
• Evaluate three objective formulations (makespan; makespan + total tardiness; makespan + average tardiness) on industrial data.
• Demonstrate improvements over simpler reference plans and analyze optimality on small to medium instances.

提出的方法

• Formulate a compact constraint programming model that captures fixed routings, designated machines, and site-specific rules.
• Model explicit resource calendars including weekends and planned maintenance.
• Derive sequence-dependent cleaning times from site tables to govern campaign sequencing.
• Implement the model using an open-source constraint solver.
• Evaluate on deterministic snapshots from a solid oral dosage facility across three objective formulations.
• Compare against reference plans that solve a simplified variant without the site rules.

实验结果

研究问题

• RQ1 How does the data-driven constraint programming framework perform on a pharmaceutical job shop with fixed routings and site calendars across multiple instance sizes?
• RQ2 What is the impact of site-specific rules (calendars and cleaning times) on makespan and tardiness relative to simplified reference plans?
• RQ3 How do three objective formulations (makespan; makespan plus total tardiness; makespan plus average tardiness) compare in terms of feasibility, optimality, and late-job counts?

主要发现

• Makespan reductions of 88.1%, 77.6%, and 54.9% are observed across the three industrial instances as compared to reference plans.
• Total tardiness reductions of 72.1%, 58.7%, and 18.2% are achieved for the corresponding instances.
• Composite objectives further decrease late job counts with negligible makespan change on the smallest instance and a modest increase on the largest instance.
• Optimality is proven on the small case, with relative gaps of 0.77% and 14.92% on the medium and large cases under a fixed time limit.