| Auction, as a unique price formation mechanism, an effective resource allocation method and an important tool for theoretical analysis, has had a good prospect of being widely used in the market over the past few years. In the field of economy, many newly emerged markets can be well interpreted through auction theories. With the intensified globalization and integration of economy, increasingly specialized and minute division of labor as well as enterprises’dedication to fostering core competitiveness, logistics has begun to be separated from traditional commodity circulation and become a crucial part for the operation of national economy. With the emergence of modern logistics represented by third-party logistics, logistics markets become more and more prosperous, where the selection of logistics suppliers is the key for the success of logistics transactions and a crucial part for making good preliminary preparation and avoiding outsourcing risks. As to the application of multi-attribute decision-making and auctions in selecting logistic suppliers, this paper focuses on comparatively studying these two methods and utilizing them as mutual references to improve the use of multi-attribute auction models in logistics transactions by introducing theories and methods about multi-attribute decisions and releasing some constraints in multi-attribute auction models, for the purpose of enhancing the reliability and rationality of these models in the selection of logistics suppliers.At present, to comprehensively evaluate logistics suppliers by multi-attribute decision-making theories is a mainstream research method for the selection of these suppliers. As crucial trading methods conforming to principles of equality, fairness and openness, auctions have been utilized more and more widely in trading markets of logistics services with their increasingly extensive involvement in market transactions. Some scholars have come to realize that modern auction theory based on hypothesis of bounded rationality, utility theory and depiction of asymmetric information by possibility distribution and particularly multi-attribute auction theory are quite suitable for the research on the selection of logistics suppliers. Multi-attribute decisions and multi-attribute auctions have their respective advantages and deficiencies. This paper comparatively studies the differences and similarities of these two methods, which is favorable for their integrated use, so the selected topic is of great theoretical significance. To be specific, this paper primarily reaches following conclusions:First of all, aiming at the selection of third-party logistics suppliers, this paper introduces the research background of this topic. In combination with common analysis methods for selecting logistics suppliers, it summarizes domestic and foreign research status, revealing theoretical and practical significances of this research. Besides, it sums up the research ideas of this paper, elaborating the research contents of all chapters and innovations of this dissertation. Additionally, this paper introduces theories used in analyzing the selection of third-party logistics suppliers covering auctions, multi-attribute decisions and logistics outsourcing, etc. Furthermore, the author constructs basic schematic diagrams for classifying different auction methods and deduces equilibrium bidding strategies for four categories of traditional auctions according to revenue equivalence principles as specified in auction theories.Secondly, this paper makes a summary of basic concepts and theories involved during logistics outsourcing. Specifically, it demonstrates transaction cost theory, core competitiveness theory, principal-agent theory and resource-based theory concerning logistics outsourcing after acquiring a clear understanding on the concepts&features of logistics outsourcing&logistics services and having an overview of foreign and domestic research status for logistics outsourcing. Besides, the author deeply analyzes the benefits the logistics outsourcing has created for corresponding enterprises from the perspective of business costs, core competitiveness, market response and risks encountered by enterprises, etc. Meanwhile, this paper discusses basic principles and general steps that shall be followed in the selection of third-party logistics suppliers, establishing an index system for evaluating third-party logistics suppliers.Thirdly, this paper summarizes multi-attribute auction theory and designs an optimal multi-attribute auction mechanism for maximizing social benefits. To be exact, it sums up domestic&foreign research findings on multi-attribute auction theory and the utilization of the theory in examining the selection of third-party logistics suppliers. It also demonstrates the superiorities of auction theory particularly multi-attribute auction theory in incentive compatibility reflected from the formation of effective market competition, revelation of decision-makers’ preferences and design of asymmetric information, analyzing the rules of distribution and payment of multi-attribute sealed-bid auction under the first-price and second-price rules. Then, the problems on selection of third-party logistics suppliers are abstractly converted into a specific model for multi-attribute reverse auctions of all quality attributes in single-round sealed-bid auctions. Besides, the author designs an optimal auction mechanism in which the objective function is maximum expected social surplus pursuant to second-price rules. Related simulation results verify the rationality of this auction mechanism.Fourthly, this paper summarizes multi-attribute decision-making theories and improves TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) by introducing Mahalanobis distance. The selection of logistics suppliers, in essence, is a multi-attribute decision-making problem, while multi-attribute decision-making approach is widely used in this field. This paper reviews the research findings on multi-attribute decision-making theories in selecting logistics suppliers from the perspective of constructing index systems and choosing evaluation methods. Based on this, it sums up the characteristics and demerits of typical multi-attribute decision-making approaches, which provides references for weight combination or combination of evaluation results in single multi-attribute evaluations. In consideration that TOPSIS is extensively applied in multi-attribute decision-making, has intuitive geometric principles and realizes simple calculations, this paper analyzes the objective fact that the Euclidean distance is ineffective under the condition that attribute indexes are linearly correlated, investigates the relationships between the formulas for Mahalanobis distance and Euclidean distance and demonstrates that Euclidean distance is an exceptional case of Mahalanobis distance. Furthermore, it eliminates dimensional and order of magnitude effects among attribute indexes by Mahalanobis distance as well as the features of related linear problems, based on which the distance calculation by TOPSIS under more general conditions is expanded and the application of TOPSIS is promoted. The analysis results suggest that dimensional and order of magnitude effects among attribute indexes and linear relationships between attribute indexes are eliminated by Mahalanobis distance, which efficiently improves scientific and reasonable application of TOPSIS in comprehensively assessing logistics suppliers.Fifthly, this paper comparatively analyzes two fundamental methods including multi-attribute decision-making and multi-attribute auction. By introducing multi-attribute decision-making approach (principal component analysis), it improves the deficiencies of traditional multi-attribute auction models. To be specific, it conducts a comparative analysis on the differences and similarities between multi-attribute decision-making and multi-attribute auction from the perspective of theoretical analysis and practical application. Hypothetical conditions are released from traditional multi-attribute auction models, namely there are no linear associations between attribute indexes. The quality attributes of all dimensions are congregated as low-dimension quality attributes (including2-3principal components) with no linear correlations by reducing dimensions through principal component analysis. The substitution of new principal components for original quality attributes in utility functions, on one hand, eliminates the linear correlations between original attribute indexes and ensures the quasi-linear features of utility functions; on the other hand, reduces the dimensions of index system on the premise of retaining a considerable amount of information (over85%) and improves the intuitiveness of acknowledging things. The simulation results show that improved multi-attribute auction model based on principal component analysis is highly objective and effective in selecting logistics suppliers, which indicates that it is necessary and reasonable to integrate the use of multi-attribute decision-making approaches and multi-attribute auction theories.
|
PDF Full Text Request