This thesis presents a two-stage stochastic optimization model developed for ASML to address the challenge of workforce planning under demand uncertainty. The objective is to determine a cost-minimizing distribution of engineering manpower across global customer-support regions while ensuring that regional demands can be met across multiple uncertain future workload scenarios. The problem is structured into two stages to reflect organizational realities. The first-stage decision involves the long-term allocation of new engineering hires across regions, which must be made before demand uncertainty is realized. The second-stage decision determines the optimal cross-regional support strategy—an operational recourse action where engineers are dynamically deployed to meet realized demand scenarios. The model is formulated as a deterministic equivalent linear program that minimizes the total expected cost. This cost comprises first-stage regional labor costs and expected second-stage operational costs, which include travel expenses for cross-regional support and penalties for manpower over-utilization. Key operational constraints are incorporated, including strict demand fulfillment requirements, hard capacities on deployable regional manpower, and target utilization rates to prevent excessive workload and preserve work-life balance. Additionally, the model respects baseline manpower distributions to focus purely on allocating new budget capacity rather than reallocating existing employees. By decoupling long-term capacity planning from short-term operational recourse, this model provides ASML management with an actionable, data-driven framework to evaluate hiring strategies, assess trade-offs between local hiring and cross-regional deployments, and optimize manpower investments in an uncertain semiconductor manufacturing environment.

A Cost-Optimization Model for Workforce Planning Under Demand Uncertainty

DE MARINIS, MICHELE
2025/2026

Abstract

This thesis presents a two-stage stochastic optimization model developed for ASML to address the challenge of workforce planning under demand uncertainty. The objective is to determine a cost-minimizing distribution of engineering manpower across global customer-support regions while ensuring that regional demands can be met across multiple uncertain future workload scenarios. The problem is structured into two stages to reflect organizational realities. The first-stage decision involves the long-term allocation of new engineering hires across regions, which must be made before demand uncertainty is realized. The second-stage decision determines the optimal cross-regional support strategy—an operational recourse action where engineers are dynamically deployed to meet realized demand scenarios. The model is formulated as a deterministic equivalent linear program that minimizes the total expected cost. This cost comprises first-stage regional labor costs and expected second-stage operational costs, which include travel expenses for cross-regional support and penalties for manpower over-utilization. Key operational constraints are incorporated, including strict demand fulfillment requirements, hard capacities on deployable regional manpower, and target utilization rates to prevent excessive workload and preserve work-life balance. Additionally, the model respects baseline manpower distributions to focus purely on allocating new budget capacity rather than reallocating existing employees. By decoupling long-term capacity planning from short-term operational recourse, this model provides ASML management with an actionable, data-driven framework to evaluate hiring strategies, assess trade-offs between local hiring and cross-regional deployments, and optimize manpower investments in an uncertain semiconductor manufacturing environment.
2025
A Cost-Optimization Model for Workforce Planning Under Demand Uncertainty
Cost-Optimization
Workforce Planning
Demand Uncertainty
File in questo prodotto:
File Dimensione Formato  
THESIS.pdf

Accesso riservato

Dimensione 2.65 MB
Formato Adobe PDF
2.65 MB Adobe PDF

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/115832