Interesting Operations Problems in Mobile Money at the Base of the Pyramid

Traditionally, people living at the base of the pyramid have lacked access to banks and other formal financial systems and products.  However, most people have mobile phones, or at least SIM cards.  Recently, the development of mobile money systems has allowed these people to deposit, withdraw, save, and electronically send money using their mobile phones.  The adoption of mobile money has been instrumental in lifting families out of poverty by offering clients mechanisms to save and transfer digital currency, which has helped to smooth financial shocks that would otherwise have been devastating. 

However, adoption of mobile money has lagged behind the hopes of advocates.  One hypothesis is that customers are wary of embracing the technology wholeheartedly due to low service levels at deposit and withdrawal points: when they want to make deposits or withdrawals, the agents at these points do not have enough physical or electronic cash to complete the transaction.  Mobile money agents are entrepreneurs who carry physical and electronic cash to make deposits on behalf of customers, and work on commission from platform operators.  They must decide how much money to carry, and how to divide this between physical and electronic cash.

We examine the "Agent mobile money inventory problem" through two papers.  First, we model the agent's profit-maximizing decision, and develop a heuristic that performs near optimally.  We then implement this heuristic in a field experiment in Tanzania in the form of offering suggestions to agents daily in terms of how much to carry each day.  While daily suggestions made an impact, there is still a large optimality gap.  Thus, we share research-in-progress and conclude by proposing more efficient operational models that could better coordinate the supply chain, and increase service levels of agents.

Jason Acimovic is an Associate Professor of Supply Chain Management at The Smeal College of Business at Penn State University.  Jason’s research applies network optimization and operations research to non-traditional supply chains, with a specific focus on the management of inventory.  Recent projects include: developing optimization-based metrics for disaster relief organizations, empirically quantifying the impact of transportation capacity in disaster relief, creating (and field-testing) heuristics for mobile money agents in Tanzania, and improving demand fulfillment decisions for omni-channel and online retailers.  His work has appeared in journals such as the ASA’s Journal of Quantitative Analysis in Sports, Manufacturing & Service Operations Management, and Journal of Operations Management.