Lot Sizing and Scheduling Optimization Techniques

Lot sizing and scheduling optimisation addresses the twin challenges of deciding how much to produce or order (lot sizing) and when to process those lots on available resources (scheduling). These problems are typically formulated to minimise total cost, which may include production, inventory holding, setup and backorder penalties, while respecting capacity, demand and timing constraints. Classical exact approaches rely on mixed-integer programming models, branch-and-bound algorithms and cutting-plane methods that exploit valid inequalities to tighten relaxations. In parallel, heuristic and metaheuristic methods—such as genetic algorithms, simulated annealing and neighbourhood search—offer scalable solutions for large industrial instances, albeit with no guarantee of global optimality. Hybrid or matheuristic frameworks combine exact solvers with local search to strike a balance between solution quality and computational effort. Recent advances also explore integrated models that couple lot sizing and scheduling decisions directly, as well as the incorporation of machine learning techniques to approximate or learn capacity consumption and setup dynamics. Application domains span from plastics injection moulding and beverage production through to pharmaceuticals and biopharmaceutical multi-site manufacturing. The global significance of these methods is underscored by their direct impact on supply-chain agility, cost efficiency and sustainable resource use in modern manufacturing networks.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

A novel mathematical model for multi-level pharmaceutical tablet manufacturing has extended capacitated lot-sizing and scheduling by integrating deterministic shelf-life rules. The proposed formulation captures linked lot sizes, backorders and integrated first-expire-first-out logic, and is solved using exact optimisation for small instances and bespoke heuristics for larger problem sizes, yielding improved trade-offs between cost and shelf-life compliance. In the plastic injection moulding industry, a hybrid optimisation approach has been introduced that blends metaheuristics with list-based scheduling. Two metaheuristic variants—stochastic descent and simulated annealing—are embedded within a generic algorithmic structure and demonstrate superior performance over existing scheduling policies, achieving high-quality solutions rapidly and adapting flexibly to varied machine and resource configurations. On the theoretical front, research into valid inequalities for two-period relaxations of big-bucket lot-sizing problems has deepened understanding of the polyhedral structure underpinning relaxations with zero setup times. Families of facet-defining inequalities and associated exact and heuristic separation algorithms have been developed, significantly enhancing lower bounds and accelerating branch-and-cut solution times on benchmark instances.

Lot Sizing and Scheduling Optimization Techniques publication trend

The graph below shows the total number of articles in lot sizing and scheduling optimization techniques across all publications each year (not limited to Nature Index journals).

Technical terms

Lot sizing: The decision problem of determining production or order quantities over a planning horizon to meet demand at minimal cost while respecting capacity and setup constraints.

Scheduling: The assignment of lots or jobs to resources over time, ensuring that sequence-dependent setups and timing constraints are satisfied.

Mixed-integer programming (MIP): An optimisation framework in which some decision variables are required to take integer values, enabling precise modelling of discrete choices such as production setups.

Metaheuristic: A high-level, problem-independent algorithmic strategy—such as genetic algorithms or simulated annealing—designed to explore solution spaces efficiently for near-optimal solutions.

Valid inequality: A linear constraint added to an optimisation model that excludes infeasible fractional solutions without cutting off any feasible integer solutions, thereby strengthening the linear relaxation.

References

1. Lot-Sizing and Scheduling for the Plastic Injection Molding Industry—A Hybrid Optimization Approach. Applied Sciences (2021).
2. Valid inequalities for two-period relaxations of big-bucket lot-sizing problems: Zero setup case. European Journal of Operational Research (2018).
3. Integrated shelf-life rules for multi-level pharmaceutical tablets manufacturing processes. International Journal of Production Research (2024).
4. Constraint learning approaches to improve the approximation of the capacity consumption function in lot-sizing models. European Journal of Operational Research (2025).
5. A new lot sizing and scheduling heuristic for multi-site biopharmaceutical production. Journal of Heuristics (2017).