An absorbing Markov chain for accounts receivables

This paper presents a case study that examines the nature and extent of timely payment of customer transactions on account of a Lean manufacturing organization. A discrete-time absorbing Markov chain model is applied as a tool for analyzing five transient states (new transaction, one-month overdue, two-months overdue, three months overdue, and four months overdue) and two competing absorbing states (paid in full and bad debt write-off). By employing a first-order Markov chain, transient states in the collections process are modeled by periods beginning with the completion of a new transaction. Ordinary Markov chain calculations are performed to determine the expected status of accounts receivables collections.