A leagile inventory-location model: formulation and its optimisation

Leagility is defined as the capability of deploying lean and agile paradigms simultaneously. This paper uses transshipments (i.e. monitored movements of stocks among locations at the same echelon) as a strategic tool to achieve leagility in an inventory-location model. Authors have coined a new term 'leagile inventory-location model (LILM)' that addresses leagility by managing inventory at numerous locations. In this paper, LILM is first formulated as a non-linear integer programme and then solved in real time with the aid of genetic algorithm (GA), genetic algorithm with chromosome differentiation (GACD) and virus-evolutionary genetic algorithm (VEGA). These algorithms are tested on a simulated 88-retailer problem with rigorous analyses of the results. It is found that, in three out of 13 instances, total costs obtained by VEGA is minimum; while in the remaining, GACD outperforms both the VEGA and the GA. Conversely, performance of GA dominates in terms of CPU time. Impact of various parameters on the results is also scrutinised and reported accordingly.