Module selection in cash flow logistics: a special location-assignment-routing problem

This paper investigates the optimisation of cash logistics network design for a set of self-service banking devices. Two types of modules are used: (i) Automated Teller Machines (ATMs) which only dispense cash, and (ii) Cash Recycler Systems (CRSs) which can both dispense and accept cash. Although CRS units involve higher acquisition costs than ATMs, their deployment can significantly reduce operational cash logistics costs when installed at suitable locations. Independent ATM Deployers (IADs) are emerging companies that own and operate ATMs independently of banks or financial institutions. To model and analyse this problem, we develop a specialised location–inventory–routeing problem with time windows and module selection. A novel metaheuristic solution approach based on invasive weed optimisation is proposed. The applicability of the model is demonstrated through a real-life case study conducted with an IAD company. The results indicate that installing CRSs is particularly advantageous at locations that are geographically remote from other devices but exhibit sufficient cash return rates. Moreover, CRSs are shown to be a preferable option in areas close to depots with high return volumes, as well as in distant locations even when return rates are relatively moderate.