پاسخ موثر به RFQs و توسعه تامین کننده: دیدگاه یک تامین کننده

Considering multiple attributes while evaluating request for quotes (RFQs) responses from suppliers is gaining significant importance in industrial procurement. While price has traditionally been the most important factor in evaluating RFQ responses, incorporation of non-price attributes such as quality and delivery performance is becoming essential and critical. Research on multi-attribute RFQs has received significant attention in an auction format with models addressing issues relating to auction mechanism design, winner determination, and auction dynamics, primarily from a buyer's perspective. There have been few approaches, if any, that have investigated the issue of response to multi-attribute RFQs from a supplier's perspective, which is the focus of this paper. Such an approach will assist a supplier in effectively responding to RFQs, thereby maximizing the likelihood of winning future contracts. It also indirectly assists in supplier development and helps foster competition among suppliers, which benefits both the buyers and the suppliers. We develop mathematical models that address this important issue and demonstrate their usefulness through an illustrative dataset.

Evaluating multidimensional request for quotes (RFQs) is an important aspect of industrial procurement. Incorporating multiple dimensions into an RFQ is replacing the traditional price-based procurement. While price-driven procurement strategies may be appropriate for commodity-related transactions, procurement of complex products and systems requires consideration of attributes in addition to price. Some of the non-price attributes considered by industrial firms while selecting suppliers are delivery, quality, design and worldwide supply capability, and cost reduction performance. Vijayan (2000) discusses the importance of considering multiple attributes for procuring custom-engineered products. Companies such as Whirlpool, Arvin Meritor, Boeing, and Northern Telecom use both price and non-price attributes in their RFQs. Wise and Morrison (2000) observe that price-based selection hinders participation of high-quality, innovative suppliers in the procurement process. These studies suggest that both the buyers and the suppliers need decision support systems that consider a variety of product- and supply-related attributes for effectively engaging in RFQ-based procurement. The problems of RFQ-based procurement have primarily been analyzed from a reverse auction perspective, in which a buyer selects one or more suppliers who meet certain price and non-price requirements (Teich et al., 2004 and Teich et al., 2006). The related literature in this area has mainly focused on auction mechanism design, auction dynamics, and winner determination (in auctions) from a buyer's perspective. The supplier's problem of effectively responding to multi-attribute RFQ has not been adequately addressed in the extant literature. Addressing the supplier's perspective is important for both the buyers and the suppliers for the following reasons. First, suppliers can benefit immensely from an approach that identifies how to respond to multi-attribute RFQs so as to maximize the likelihood of winning future contracts. Second, it assists the buyer to foster competition among suppliers, which indirectly helps in supplier development. A buyer can provide feedback to suppliers by sharing the information on attribute values for past winning quotes, which can help the suppliers to discern the buyer's requirements and to develop the necessary capabilities to successfully respond to future RFQs. We posit that in an RFQ-based procurement context, such an indirect “indicative mechanism,” i.e., sharing information on past winning quotes with the suppliers, entails little or no direct investments by the buyer compared to the traditional supplier development initiatives in a supply chain involving investments for cost reduction, quality improvement, lead-time reduction. Thus, by inducing the suppliers to participate in the RFQ-based procurement process, the buyer can pursue the goals of continuous supplier development without direct investments. Hax and Majluf (1988) describe strategy as “a pattern of actions that emerge from the past decisions of the firm that purposefully manages change.” Thus, historical pattern of winning quotes across multiple attributes can reflect the sourcing strategy of the buyer, which the supplier can and should attempt to infer and use to reconfigure future quotes. And, it is in the interest of the buyer to provide information on the winning quotes to the suppliers. It can be expected to foster competition among suppliers, potentially leading to higher levels of performance across the attributes of interest to the buyer, since the suppliers impelled by self-interest can be expected to actively engage in performance improvement efforts. This discussion underpins the problem studied in this paper. We develop mathematical models that address a supplier's problem of effectively responding to multiple-attribute RFQ in industrial procurement contexts in durable goods industry. The main contributions of the suggested models are developing a composite measure of attributes that represents the delivered value per dollar to the buyer, identifying alternative levels for various attributes (price, quality, and delivery) that a supplier can quote to maximize the likelihood of winning future contracts, and effectively integrating inexact and inferred buyer's preferences into the decision process for both single- and multiple-winner cases. It is observed that reverse auction models typically require the buyer to specify the exact preference function a priori to potential suppliers ( Bichler, 2000). In industrial procurement, especially in durable goods industry, it is an arduous task for the buyer to specify the exact relative preferences for each attribute because procurement decisions often involve a group of procurement managers. Identifying a precise ‘group preference function’ is impracticable as the group membership often varies across procurement cycles. Due to these and other issues, it is often more practical for the buyer to specify acceptable threshold levels for various attributes and inexact preferences pattern that capture the relative importance among attributes and not the exact weights. This is the approach that we use in our model development. We next discuss the problem setting, proposed models, and the practical usefulness of the models.

Responding to multi-attribute RFQs requires a supplier to accurately infer the unknown attribute preference weights of the buyer. In this paper, we proposed a set of multi-attribute models for developing effective quoting strategies from a losing supplier's perspective based on the past winning quotes. Although, we considered cost, quality, and delivery performance in our formulation, the approach can be extended for additional attributes. Our approach provides an innovative way of responding to RFQs in multi-attribute bidding with several advantages over the existing methods. These include capturing the preference weights of the buyer with the buyer's minimal involvement, considering the tradeoff interrelationships among various attributes, and identifying alternative target attribute levels for the unselected supplier in order to maximize the likelihood of winning the future supply contract. The investigation has significance for strategic procurement since it extends the notion of supplier development to the RFQ-based procurement context. Supplier development is a key activity in sourcing. It has been traditionally viewed as a buyer-driven initiative, which is implemented through a series of collaborative efforts that entail investments by the buyer. Our study suggests an alternative indirect approach to supplier development. The buyer's procurement priorities can be indicated to the suppliers in terms of information on past procurement decisions and inexact preferences for various attributes. This information could be utilized by suppliers in identifying the options to effectively respond to RFQs in the future procurement cycle. This sequence of events can be expected to lead to development of long-term supplier capabilities through a number of short-term operational improvements that are carried out in each procurement cycle. We consider both single-winner and multiple-winner cases in each procurement cycle. The illustrative example demonstrates the usefulness of the model. There are several opportunities for further research to improve the scope and applicability of the model. The proposed models mainly address the short-term concerns of the unsuccessful supplier to succeed in the next procurement cycle. It does not address the issue of how to proactively modify the strategy for subsequent periods after the supplier becomes successful. Strategic sourcing goes beyond the one time transaction between a buyer and a supplier. It involves issues such as long-term partnership for building mutual competency, innovation, and cost reduction that are difficult to include in our model. Also, our models assume certainty of supplier performance. The model needs to be improved for considering the above-mentioned ordinal attributes and performance variances. Another possible extension could be in simultaneously solving for the optimal levels of attributes under supplier specified price, quality, and delivery restrictions. In such a model, the sensitivity of various parameter coefficients and right-hand side values of constraints may provide interesting managerial insights relating to the preferences of the buyer and limitations of the supplier. Selecting the optimal set of suppliers can also be an area for future investigation.