Sequential Hiring of Contingent Workers Through Learning-Based Optimization
Researchers have developed a new learning-based optimization policy called DR-UCB to address the sequential hiring of contingent workers. This policy aims to maximize cumulative profit by managing an active team of a fixed size while learning worker productivity over time. DR-UCB accounts for the costs associated with replacing workers and potential delays in hiring due to external commitments or onboarding processes. The policy makes sequential decisions on when to initiate workforce changes and which workers to replace and hire, demonstrating superior performance compared to benchmark policies in numerical experiments. AI
IMPACT This research introduces a novel approach to workforce management that could improve efficiency in industries relying on contingent labor.