There are a number of reasons responsible to determine the Returns policies. A well-structured overview of types of returns policies and reasons for their use is provided by Padmanabhan and Png[49]. Chu [50] has cited two reasons for ordering to satisfy the demand. One reason is, when demand is high it allows a manufacturer to credibly signal demand information to the distributor, inducing him to order enough product. Another reason is to reduce the distributor‘s overage cost, inducing him to order more. Hence a much needed return policy is required to avoid the distributor to order less than in the integrated channel since his margin is smaller than the integrated channel. According to Pasternack[51] if returns
parameters are properly chosen it can induce the distributor to order an optimal amount for the channel. The distributor and manufacturer can agree upon a pair that is Pareto optimal within this set of parameters, a large number of papers with similar approach as Pasternack [51] adopted for single-product returns policies.
The basic approach is to find a returns policy structure with a number of parameters like wholesale price, returns price, return allowance, etc and then finding sets of these parameter values to induce the distributor to make order decisions that optimize the integrated channel. Finally one of the sets is selected so that both parties have larger expected profit. Many supply contracts are mathematically equivalent to returns policies were discussed by Lariviere,[52]. A good number of papers using this approach under various policies and models including Kandel[53], Emmons and Gilbert [54], Donohue [55], Taylor [56],Tsay[57] and Brown and Lee [58] are available. A different approach considering the issue of risk is put forth by Webster and Weng[59]. They have evaluated the returns policies that increase the retailer‘s expected profit while ensuring the manufacturer‘s worst case profit will always be at least as large.
Ferguson et al. [60] examine a situation in which consumers may return the products to the retailer with no functional or cosmetic defect. A rebate mechanism is developed by them to attract the retailer to increase her effort to reduce the number of false returns. By considering the single-product case, they show that the rebate mechanism is Pareto optimal. Focusing on the relationship between performance of companies in terms of managing product returns and customer loyalty, Ramanathan [61] studied in the context of the business-to-consumer (B2C) segment of electronic commerce,. The analysis presented mainly contributes to the literature by providing a risk perspective to existing studies. They have analyzed the relationship between performance of companies in handling product returns and firm performance. Their analysis considers only ease of returns/ refunds as the indicator.
An analytical model for the use of e-marketplace in a supply chain is provided by Choi,T.M. et al[62]. They have considered a two-echelon supply chain with a single manufacturer who supplies a single item to a distributor and the manufacturer with a buyback policy. Under the returns policy, the distributor can return any unsold product to the manufacturer for a partial re-fund after the selling season is over.
2.6 Reverse distribution
Reverse distribution is the compilation and transportation of used products and packages for different reasons. Reverse distribution can be executed through the original forward channel, through a separate reverse channel, or through combinations of the forward and the reverse channel. Guiltinan and Nwokoye[63] provided one of the first analyses of reverse distribution networks according to the actors involved. Pohlen and Farris [64] claim that, the reverse channel depending on individual channel members' functions and ability to perform recycling or remanufacturing tasks may take several different forms. A major issue in reverse distribution systems is the question if and how forward and reverse channels should be integrated. In order to set up an efficient reverse distribution channel, decisions have to be made with respect to:
The actors in the reverse distribution channel.
The actors may be members of the forward channel (e.g. traditional manufacturers, retailers, and logistics service providers) or specialized parties (e.g. secondary material dealers and material recovery facilities). This distinction sets important constraints on the potential integration of forward and reverse distribution.
The functions have to be carried out in the reverse distribution channel.
The possible functions in the reverse distribution channel are: collection, testing, sorting, transportation, and processing. For finding out suitable locations for these functions, a distribution network is to be designed. One important issue is the location of sorting and testing within the network.
Timely testing might save transportation of useless returns. On the other hand, sophisticated testing might involve expensive equipment which can only be afforded at a few locations. Decentralized testing is therefore typically re-stricted to a rather rough, preliminary check. Sorting of a return stream into different reusable fractions (e.g. in household waste collection) might be less expensive at an early stage close to collection.
However, subsequent handling costs may increase and transportation ca-pacity utilization may decrease for early splitting into distinct streams.
Customer ability (and willingness) to partly carry out the sorting function is another aspect to be considered Jahre, [65].
The relation between the forward and the reverse distribution channel.
Recycling can often be described as an open-loop system, i.e. the products do not return to the original producer but will be used in other industries.
Possibilities for integration of forward and reverse distribution are scant as the actors differ in both channels. Remanufacturing and reuse often lead to closed-loop systems: the product or packaging returns to the original producer.
Reverse distribution may either take place through the original network directly, using traditional middlemen or through specialized logistical providers. Even if the same actors are involved, integration of forward and reverse distribution may be difficult at the routing level since collection and delivery may require different handling.
Fig. 1.7 shows a framework for reverse distribution combining the forward flow from producer to user, and the reverse flow from user to producer.
Within this framework, Operational Research methods have been applied to study reverse flow networks. The focus has mainly been on network design issues. We describe models for the separate reverse flow problem in Section 3.1. Models partly using the original forward network for the reverse distribution are described in Section 3.2.
• Separate modelling of reverse flow
Several authors have proposed modifications of traditional facility location models, Mirchandani and Francis, [66] for the design of reverse distribution networks. One special characteristic to be taken into account is the convergent structure of the network from many sources to few demand points, Ginter and Starling, [67].Such 'many-to-few' problems have also been studied in the hazardous waste disposal literature. Batta and Chiu[68], has considered the problem of determining optimal paths for routing an undesirable vehicle on a network embedded on an Euclidean plane. His work involves in finding out a path that minimizes the weighted sum of lengths over which this vehicle is within a threshold distance lambda of population centers. Erkut [69] did use the decision support systems that provide sound directions for transportation of hazardous materials. By contrast, traditional location models typically consider a divergent network structure from few sources to many demand points.
Another peculiarity of reverse distribution networks is their high degree of uncertainty in supply both in terms of quantity and quality of used products returned by the consumers. Both are important determinants for a suitable network structure since, e.g., high quality products may justify higher transportation costs (and thus a
more centralized network structure), whereas extensive transportation of low value products is uneconomical. Moreover, end-markets for recovered products may not be well known, exposing network planning in this context to even more uncertainty.