Christos Zikopoulos

Impact of uncertainty in the quality of returns on the profitability of a single-period refurbishing operation

This paper investigates how the profitability of reuse activities is affected by uncertainty regarding the quality of returned products. Specifically, we examine a reverse supply chain consisting of two collection sites and a refurbishing site, which faces stochastic demand for refurbished products in a single-period setting. The quality of returns (refurbishing yield) becomes known only after the transportation of the products to the refurbishing site. We prove that the expected profit function has a unique optimal solution (procurement and production quantities) and we derive the conditions under which it is optimal to use only one of the collection sites. The analysis is supported by numerical results which provide insights regarding the effect of the uncertain yields at the two collection sites and their correlation on optimal decisions and system profitability.

On the attractiveness of sorting before disassembly in remanufacturing

We examine the attractiveness of simple sorting procedures characterized by limited accuracy just before disassembly and remanufacturing of used products. That type of quick sorting is often made possible through the installation of simple electronic devices in new products, which record basic usage data and provide information about the remanufacturability of the product without the need for its disassembly. We study a two-level reverse supply chain with remanufacturing and we concentrate on the single-period setting. There is uncertainty about the remanufacturability of used products and we derive the conditions under which quick sorting is economically justifiable. We show that the economic attractiveness of sorting depends on the costs of transportation, disposal, disassembly, the cost and accuracy of the sorting procedure and the expected quality of the returned items.

On the effect of quality overestimation in remanufacturing

We study a simple reverse supply chain consisting of a remanufacturing facility and a number of independent locations where used products are returned by the end-users. At the collection locations, the returned products are graded and classified based on a list of nominal quality metrics provided by the remanufacturer. It is assumed that this classification is subject to errors; specifically, the returns condition is overestimated because of a stochastic proportion of returned units which are classified in classes corresponding to better quality than the actual. The scope of the paper is to study how these classification errors affect the optimal procurement decisions of the remanufacturer as well as the associated profit for the cases of both constant and stochastic demand in a single-period context. Moreover, in the former case we study the impact of these classification errors on profit variability. The quantification of the impact of quality overestimation provides intuition on the value of reliable classification and on the extent of the necessary investments and initiatives to improve classification accuracy.

Optimal procurement and sampling decisions under stochastic yield of returns in reverse supply chains

We study the problem of optimizing the sampling and procurement decisions in a remanufacturing system under stochastic yield of returns in a single-period setting. Returned products are characterized by uncertainty regarding their ability to be successfully remanufactured. This uncertainty is formulated as a variable probability of each returned unit in a batch to be remanufacturable (returns yield). We study the impact of returns yield on the optimal procurement policy and the benefits of sampling inspection of returns prior to the procurement decision. It is shown that sampling inspection improves the procurement decisions since it allows the Bayesian updating of the prior information regarding the returns yield. We derive analytical expressions for the determination of the economically optimal procurement quantity and structural properties that facilitate the optimization procedure and provide useful insights. The determination of the economically optimal sample size is also discussed and the benefits of sampling are illustrated through numerical examples.

Joint optimization of manufacturing/remanufacturing lot sizes under imperfect information on returns quality.

The increased quality volatility that characterizes returned used products in reverse supply chains, has led both academic research and industrial practice to support the establishment of, preferably fast and inexpensive, procedures for the quality assessment of returns. To this end, an initial classification of returns is typically performed based on some easily acquirable usage information, which is inherently - yet not always perfectly - related to their quality. Despite the criticality of the returns classification process and its impact on the reverse supply chain operation and profitability, little work has been done on the determination of the optimal threshold values, used to classify returns into different quality categories, taking into account the inevitable quality assessment inaccuracies. Motivated by multiple related reports from industrial applications, in the present paper the problem of optimizing the returns classification process under partial quality information in a hybrid manufacturing/remanufacturing facility is studied. More specifically, analytical models for the joint optimization of the new and returned products lot-sizing decisions are developed, based on usage information of returns which is only indirectly related to their remanufacturability. Our analysis provides useful insights on the value of information regarding the quality of returns under an indicator-based initial classification process. Numerical evidence on the economic superiority of the proposed models under alternative quality patterns of returns is also provided.

Remanufacturing lotsizing with stochastic lead-time resulting from stochastic quality of returns

In the current paper, we model the duration of recovery of used products as a variable that depends on each unit’s quality. Because of the uncertainty related to returned units’ quality, the necessary time for the recovery of a lot is a random variable. We provide analytical expressions for the optimisation of recovery planning decisions under different assumptions regarding quality and demand characteristics. In addition, through an extensive numerical study, we examine the impact of the different parameters on the necessity to consider explicitly the stochastic nature of recovery lead-time. Moreover, we discuss the advisability of establishing procedures for the classification of returns according to their quality condition. As our findings indicate, overlooking quality uncertainty can increase related costs considerably because of poor process coordination. Furthermore, ignoring variability may result in undue overestimation of the efficiency of lot-sizing policies. On the other hand, the establishment of quality assessment procedures is worthwhile only when the stochastic behaviour of quality cannot be taken into account explicitly.

Reverse supply chains: Effects of collection network and returns classification on profitability

Used products collected for value recovery are characterized by higher uncertainty regarding their quality condition compared to raw materials used in forward supply chains. Because of the need for timely information regarding their quality, a common business practice is to establish procedures for the classification of used products (returns), which is not always error-free. The existence of a multitude of sites where used products can be collected, further increases the complexity of reverse supply chain design and management. In this paper we formulate the objective function for a reverse supply chain with multiple collection sites and the possibility of returns sorting, assuming general distributions of demand and returns quality in a single-period context. We derive conditions for the determination of the optimal acquisition and remanufacturing lot-sizing decisions under alternative locations of the unreliable classification/sorting operation. We provide closed-form expressions for the selection of the optimal sorting location in the special case of identical collection sites and guidelines for tackling the decision-making problem in the general case. Furthermore, we examine analytically the effect of the cost and accuracy of the classification procedure on the profitability of the alternative supply chain configurations. Our analysis, which is accompanied by a brief numerical investigation, offers insights regarding the impact of yield variability, number of collection sites, and location and characteristics of the returns classification operation both on the acquisition decisions and on the profitability of the reverse supply chain.

The Value of Sampling Inspection in a Single-Period Remanufacturing System with Stochastic Returns Yield

We examine a reverse supply chain consisting of a collection site, where consumers return used products, and a remanufacturing facility. Some of the returned products are transported to the remanufacturing facility in order to be remanufactured and used to satisfy the stochastic demand for remanufactured products. The quality of returns is characterized by uncertainty, and therefore, before the procurement quantity determination, the remanufacturer has the alternative to inspect a sample drawn from the collected quantity in order to evaluate more accurately returns’ quality. Using general assumptions for returns quality and remanufactured products demand distributions, we formulate the expected profit function for both sampling and no-sampling cases and we examine numerically the economic effectiveness of sampling. A key characteristic of the current paper is that returns’ yield is expressed as the probability of a unit to be remanufacturable.

Optimal Disposition of Returns Based on Inaccurate Quality Assessment Procedures

Abstract In order to obtain timely quality information of returns, remanufacturers have to rely on inexpensive, fast, yet inaccurate classification procedures. Quality assessment of returned products is usually based on the comparison of certain product characteristics or usage conditions to predetermined thresholds. Classification inaccuracies typically occur because it is very rare to identify a product attribute, which is easily evaluated and, at the same time, in perfect correlation with product quality condition. Therefore, in the majority of industrial applications, classification procedures are subject to errors, i.e., rejection of good quality units and acceptance of inferior quality units. The objective of the current research is to identify the economic trade-offs related to the determination of the acceptance / rejection criterion taking into account the inevitable classification errors, under different assumptions regarding the available quality classes and demand characteristics.

Recovery Planning with Stochastic Lot-Processing Times

Abstract Although the issue of uncertain quality of used products has a central role in reverse logistics and recovery process planning, little work has been done to study the impact of the different effort and time required for the recovery of units with different quality condition on the inventory management decisions and the resulting service levels. In the present paper we examine this issue in a multi-period horizon setting to determine the optimal parameters of a ( Q, s ) continuous review inventory management policy. Furthermore, we quantify the increase in costs because of quality uncertainty, as well as the economic benefits of taking into account this uncertainty during recovery planning.