Balancing Demand Information and Outsourcing
ANN ARBOR, Mich. -- In a manufacturing company, a manager controlling the production-inventory system must consider the effects of many unpredictable factors. Two important ones are variable customer demand and variability in the firm’s own production capacity, which impact decisions about targeted inventory levels and the procurement of additional product from suppliers. It is a tough balancing act with no easy solution, say University of Michigan Business School researchers.
Traditionally, managers have sought to protect their companies against uncertainties in demand and production by ensuring that they have safety capacity, either through outsourcing work to other vendors or relying on overtime production within their organizations. In recent times, there has been a trend toward investing in better technology systems to provide valuable advance-demand information that managers can utilize to improve the overall performance of a company’s production-inventory system. Naturally, creating sources of safety capacity and improving information capabilities both have associated costs, which managers must weigh against the final results, further complicating the decision-making process.
In an award-winning article, three University of Michigan Business School researchers -- faculty members Izak Duenyas and Roman Kapuscinski and doctoral student Xinxin Hu -- address this management conundrum. Through modeling and computational study, they formulate the optimal policy for determining a manufacturer’s outsourcing quantity versus its production quantity under conditions when capacity and demand are uncertain.
“In all cases, the optimal policy is a double-threshold structure, where the manufacturer first outsources up to a certain level and then produces up to a higher level,” said Duenyas, professor of operations management and associate dean at the University of Michigan Business School. “The type of demand information does not affect the double-threshold structure, but the values of the thresholds are determined by demand forecasts.”
If the replenishment of product levels were instantaneous, a manager could conceivably ignore all advance-demand information except for the current period, the researchers say. However, when demand uncertainty and production variability go hand-in-hand, it is more difficult to plan production levels, so better information about future demand is needed. By reducing the fluctuation in demand and/or capacity, a manager may be able to improve the overall performance of the inventory-production system.
In addition to demand and production variability, other system parameters, such as service level and utilization, can influence the individual and joint benefits of demand information and outsourcing in reducing uncertainty, they say. For example, advance demand information benefits manufacturers by helping to decrease order-up-to levels and to reduce the amount of safety stock required. But information’s value diminishes in situations where there is increased capacity variability, greater utilization and higher service levels.
Outsourcing works in exactly the opposite way, the researchers say. It is most beneficial when capacity variability, service level and utilization are increasing, thereby putting greater pressure on capacity. Under these conditions, the outsourcing option can directly and significantly decrease this pressure and, thus, becomes more valuable.
There are cases when the cost reduction gained by using both approaches at the same time far exceeds the sum of the savings from using either one individually, they say.
“When a system is pressured to keep higher inventory levels with increased demand uncertainty, a manufacturer can cut the amount of safety stock significantly because it has very good information on the current period,” said Kapuscinski, Michigan Business School assistant professor in operations management. “Furthermore, since outsourcing is also available, there is no need to carry large inventories for spikes in the future either.”
Their article, “Advance Demand Information and Safety Capacity as a Hedge against Demand and Capacity Uncertainty,” recently won first place in the Manufacturing and Service Operations Management Society’s Student Paper Competition. Hu is a doctoral student of operations management.
For more information, contact:
Phone: 734.936.1015 or 734.647.1847