Edinburgh Research Explorer

Managerial tools for mitigating inventory record inaccuracy in stochastic inventory systems

Project: University Awarded Project Funding

StatusActive
Effective start/end date1/09/14 → …
Period1/09/14 → …

Description

Accurate inventory records are essential to ensure performance of production systems. Production planning is nowadays mostly carried out via automated decision support tools. Inaccurate inventory records generate costs by hindering these tools from delivering full value. In fact, inventory control models integrated in such tools traditionally account for demand and lead-time uncertainty, while they do not account for uncertainty associated with stock levels (Chan and Wang, 2014). As pointed out in (DeHoratius and Raman, 2008) “inventory record inaccuracy is an operational problem that warrants greater attention from management scholars and managers. Innovative solutions to this problem are needed,” such as preventive or corrective techniques. Prevention can be ensured via advanced tracking technologies, such as RFID, which however can only reduce – and not eliminate – inaccuracy. For this reason, corrective actions such as inventory audits are essential in production control. In this project we aim to coordinate inventory control and inventory auditing for the first time. The practical aim of our project is to develop managerial tools to determine not only what, when and how much a manager should order or produce; but also when and where, in the production chain, she should carry out an inventory audit to reduce inventory inaccuracy and to ensure a cost-optimal control action. To assess the theoretical effectiveness of these tools, we will develop analytical models, e.g. (Kaelbling et al, 1998), addressing the open academic problem discussed in (Kök and Shang, 2014) of simultaneously scheduling production and auditing. Finally, to validate them, we will rely on an ongoing collaboration with a world-leading manufacturer of electromechanical devices. From an educational point of view, we aim to comply with the recommendations of (Sodhi and Tang, 2014) and provide a balance among different stages (awareness, framing, modeling, validation) and methodologies (case study, empirical, analytical, behavioral) in Operations Management research.