Machine Learning for Labor Optimization: A Systematic Review of Strategies in Healthcare and Logistics

Abstract

This study systematically reviews machine learning (ML) based strategies for optimizing labor resources in healthcare and logistics, with particular emphasis on outcome and domain transferability. Demand and constraints of resources vary in healthcare and fulfillment settings over time and hence make those inefficient. Predictive (supervised learning) and adaptive (reinforcement learning) tools of ML can be used to enhance labor allocation and improve operational efficiency. A systematic review covering the period from 2015 to 2024 was carried out according to PRISMA rules and details regarding labor optimization in healthcare and fulfillment centers through ML applications. Using supervised learning and reinforcement learning techniques, overall labor utilization could be enhanced by as much as a 30% reduction in wait and processing times and less costs in both healthcare and logistics industries. The transferability of an intervention across sectors appeared to be effective: the intervention was introduced successfully in one sector and boosted efficiency and quality on the other. The review suggests the adoption of applied ML for optimization in labor services across different sectors coupled with cross-industry collaboration and sharing of data between the industries along with more research into transferable models and ethical practices for maximizing workforce efficiency.