Data-Driven Optimization of Employee Allocation Management in IT Services Using the Transportation Problem Framework

Abstract

In giant organisations like Infosys, TCS and Wipro that provide IT services, the ability to place thousands of employees in globally distributed projects with different skill and cost requirements, and time constraints is a continuous requirement. The allocation practices (mostly managerial judgement or rules) of the classical allocation are highly likely to lead to the inefficient use, higher operational costs and project delays. This paper suggests the data-driven workforce allocation model, which is an extension of a classical Transportation Problem (TP) the linear programming model, which has a historical application in the optimisation of logistics. The TP formulation reduces the total cost of the assignment by modelling the employee skill groups as supply nodes and the project requirements as demand nodes to ensure complete skill-project alignment. The model was used to show a substantial increase in the precision of allocation, workforce and cost efficiency using a simulated case study to reflect operational structure of large IT companies. The comparison with Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and heuristic allocation techniques demonstrates that the TP model is always the lowest cost of allocation and maximum matching accuracy in the structured environment. The results suggest the opportunity of using operations research methods to provide scalable and transparent HR decisions, which will provide a feasible and analytically viable solution to the multi-project related staffing problems faced by modern IT companies with IT-centric products.