6/23/2021 Jenny Applequist, CSL
CSL’s Alexander Stolyar and his colleague Qiong Wang recently made a surprising discovery: lead time uncertainty in business practices can be leveraged to provide significant cost benefits.
Written by Jenny Applequist, CSL
New technologies are constantly appearing, and the natural, business, and policy environments are constantly evolving. Companies respond to this dynamic setting by diversifying their production facilities, shipping routes, and supplies to minimize costs and mitigate risk. Such continuous adaptation of complex business processes inevitably increases uncertainty, which has long been seen as undesirable. In particular, randomness of replenishment lead times—that is, the time that elapses between ordering and receiving the supplies needed to run a business—is a big piece of that uncertainty.
But what if that unpredictability is not a liability at all, but an asset?
CSL’s Alexander Stolyar and his colleague Qiong Wang recently made a surprising discovery: lead time uncertainty can be leveraged to provide significant cost benefits.
“In fact, the randomness in the lead time can be exploited to reduce the inventory cost to a potentially small fraction of the cost under constant lead times,” says Stolyar. “The idea has been rigorously substantiated for a standard single-stage model under specific assumptions—in particular, exponentially distributed lead times—and tested by some preliminary simulations for this model under more general assumptions.”
“To the best of our knowledge,” he adds, “there is no similar general finding in the literature. Our results may well rewrite some consensus views in classical textbooks.”
Stolyar and Wang both worked at Bell Labs Mathematical Sciences Research for many years, and now are both faculty members in UIUC’s Industrial & Enterprise Systems Engineering department. With Stolyar as PI, they just received a 1-year award, entitled “Turning Uncertainty into Advantage: Novel Policies that Exploit Randomness of Lead Times in Supply Chains,” from the National Science Foundation. It will allow them to build on their initial proof-of-concept work on lead time randomness, to assess the scope of their approach’s applicability, and to identify directions for improvement. The overarching goal is to develop practical strategies that will enable supply chains to benefit from lead time uncertainty, so that companies can optimize their inventory management.
The work may have broader benefits as well. “Our findings may also apply to non-business activities, such as emergency management and humanitarian operations, that often encounter high delivery time uncertainties,” notes Stolyar.
A paper that describes Stolyar and Wang’s proof-of-concept work will soon appear in Operations Research; a preprint is available online at https://arxiv.org/abs/1801.02646.