Lian Qi

The Effect of Supply Disruptions on Supply Chain Design Decisions

We study an integrated supply chain design problem that determines the locations of retailers and the assignments of customers to retailers to minimize the expected costs of location, transportation, and inventory. The system is subject to random supply disruptions that may occur at either the supplier or the retailers. Analytical and numerical studies reveal the effects of these disruptions on retailer locations and customer allocations. In addition, we demonstrate numerically that the cost savings from considering supply disruptions at the supply chain design phase (rather than at the tactical or operational phase) are usually significant.

A continuous-review inventory model with random disruptions at the primary supplier

We consider a continuous-review inventory problem for a retailer facing constant customer demand for a single product. This retailer is assumed to follow the well known and widely used order-up-to policy in making replenishment decisions, and can order from two suppliers who differ in reliability and costs. Supplier 1, the primary supplier, is cheaper, but is subject to random disruptions. Supplier 2, the backup supplier or the contingent source, is more expensive, but is perfectly reliable. If Supplier 1 is available when the inventory level at the retailer reaches the reorder point, the retailer orders from Supplier 1. Otherwise, it will wait for a while to see if Supplier 1 can recover from the disruption quickly. If so, it will still get replenishment from Supplier 1 to take advantage of its lower charge. However, the retailer will reroute to the backup supplier if Supplier 1 still does not recover from the disruption when the cap of waiting (the maximal waiting time of the retailer if Supplier 1 is disrupted) is reached. We analytically study the optimal sourcing and replenishment decisions at the retailer, and the impacts of various problem parameters on the optimal decisions. We also conduct extensive numerical experiments to compare different sourcing and replenishment decisions the retailer can make and get further managerial insights into the problem.

Analysis of a two-party supply chain with random disruptions

Abstract We study a continuous-review inventory problem of a two-echelon supply chain with random disruptions, identify properties of the optimal cost function, compare the optimal order quantity with the classical economic order quantity, analyze the sensitivity of the optimal solution, and explore the conditions under which zero-inventory ordering policy is preferred.

Personnel scheduling and supplies provisioning in emergency relief operations

The practice of emergency operations often involves the travelling of medical teams and the distribution of medical supplies. In an emergency, such as an earthquake, a medical team often has to visit various hospitals (the customers) one after another in a predetermined sequence in order to perform on-site operations that require certain amounts of medical supplies. Because of their perishable nature, the medical supplies are typically shipped in batches from upstream suppliers and kept at multiple distribution centers during the disaster relief process. The scheduling of the medical teams and the provisioning of the medical supplies give rise to a scheduling problem that involves the timely dispatching of supplies from distribution centers to hospitals in coordination with the scheduling of medical teams so as to minimize the total tardiness of the completions of the operations to be performed. We introduce a mathematical programming based rolling horizon heuristic that is able to find near optimal solutions for networks of up to 80 hospitals very fast. We also report on empirical observations with regard to the computational performance of the heuristic; we consider 5420 randomly generated test cases as well as a case that is based on an actual hospital-distribution center network in the greater New York metropolitan area. Managerial insights are drawn from numerical studies regarding the benefits of pre-positioning medical supplies at the distribution centers.

Worst-case analysis of demand point aggregation for the Euclidean p-median problem

Solving large-scale p-median problems is usually time consuming. People often aggregate the demand points in a large-scale p-median problem to reduce its problem size and make it easier to solve. Most traditional research on demand point aggregation is either experimental or assuming uniformly distributed demand points in analytical studies. In this paper, we study demand point aggregation for planar p-median problem when demand points are arbitrarily distributed. Efficient demand aggregation approaches are proposed with the corresponding attainable worst-case aggregation error bounds measured. We demonstrate that these demand aggregation approaches introduce smaller worst-case aggregation error bounds than that of the honeycomb heuristic [Papadimitriou, C.H., 1981. Worst-case and probabilistic analysis of a geometric location problem. SIAM Journal on Computing 10, 542-557] when demand points are arbitrarily distributed. We also conduct numerical experiments to show their effectiveness.