Multiskilled personnel assignment problem under uncertain demand: A benchmarking analysis.

Area of Science:

• Operations Research
• Management Science
• Industrial Engineering

Background:

• Personnel assignment problems aim to minimize staff surplus/shortage costs in service industries.
• Demand uncertainty poses a significant challenge to achieving optimal staff allocation.
• Multiskilled workforces are increasingly configured using strategies like 2-chaining to enhance flexibility.

Purpose of the Study:

• To develop and evaluate a two-stage stochastic optimization (TSSO) approach for personnel assignment under uncertain demand.
• To determine optimal multiskilling requirements minimizing training and expected staff surplus/shortage costs.
• To compare the TSSO approach with existing methods like robust optimization (RO), closed-form equation (CF), deterministic (DT), and myopic strategies.

Main Methods:

• Development of a two-stage stochastic optimization (TSSO) model for multiskilling requirements.
• Evaluation of TSSO performance against RO, CF, DT, and myopic approaches using real and simulated retail data.
• Statistical comparison of total costs across different demand variability levels.

Main Results:

• TSSO, RO, and CF consistently demonstrated the lowest average total costs across various demand variability levels.
• No statistically significant cost differences were found among TSSO, RO, and CF, identifying them as the most cost-effective approaches.
• The study provides insights into determining the optimal amount and configuration of multiskilling.

Conclusions:

• The TSSO approach offers a robust method for addressing personnel assignment problems with uncertain demand and multiskilled workforces.
• Decision-makers can use the findings to select the most suitable optimization approach based on desired conservatism and reliability.
• Future research should explore k-chaining strategies for k≥2 in personnel assignment optimization.