برنامه ریزی تعاونی در شبکه حامل اکسپرس - یک مطالعه تجربی در مورد تاثیر زمان واقعی سیستم پشتیبانی تصمیم گیری

For small transportation firms cooperation in a carrier network is a proper mean to overcome the inefficiencies from deadheading. To be successful, such a network has to secure two different but equally important aspects: the partners have to be aware of specific consolidation potentials through order exchange which is an optimization and communication problem, and, the partners have to experience incentives to contribute actively to the network which is the problem of finding a fair cost/profit allocation schema for order exchanges. In this paper we discuss the experience with the development of a Decision Support System for a specific express carrier network. We illustrate how the consolidation potentials in such a network with autonomously planning carriers can be exploited and cost effectiveness can be improved substantially through the use of a suitable distributed Decision Support System if the two success factors awareness and fairness are addressed properly.

During the last years, transportation firms are faced with increasing cost pressure and revenue erosion at the same time. Large transportation companies are able to realize a high utilization of their vehicles and acceptable operational cost by consolidating and combining orders to efficient roundtrips. Small carriers serving ad-hoc one-way shipping orders only are faced with the problem of low volume of shipments with less than truckload trips as well as dead head trips due to an imbalance among locations. This leads to cost ineffective transportation plans and/or may result in non-competitive prices. Such small-sized companies may compensate their competitive disadvantage by allying with partners to a cooperation network to establish a more profitable portfolio of orders. In such a network each partner plans his orders and his vehicle fleet independently with the option to exchange orders with partners. Partners charge their own customers based on a specific price function. The carrier operating an order for a partner receives a monetary compensation, which has been specified in a generally agreed upon compensation schema. For such a network to be successful and sustainable there are a number of critical factors: on the strategical level the choice of the right set of partners yielding enough consolidation potential as well as mutual trust is a cardinal point. On the operational level awareness is essential, i.e. existing potentials have to be detected and communicated, a task which calls for the establishment of a proper information and communication system supporting cooperative planning. Another success factor is the establishment of a compensation schema which puts incentives to both partners involved in exchanging orders and which is considered to be fair by all network partners. The issue of reducing transportation cost by collaboration and forming alliances has been investigated in the scientific literature with respect to different aspects like different transportation markets or modes i.e. airfreight [7], shipping [1], trucking [8], intermodal freight transportation [16], supply chain management [18] as well as decision level, i.e. from evaluating strategic aspects in alliance formation [2] to pricing/revenue management and allocation of cost benefits [4], [10], [11] and [12]. Most developments concern design questions and propose models which are based on theoretical foundations stemming from game theory, combinatorial auctions and network flow. Refs. [10] and [13] present approaches for a problem where intra-enterprise in and outsourcing decisions on bundles of logistic services i.e. bundles of transportation requests from customers in a profit center structured forwarding company have to be made. In both approaches the cost difference through exchanging requests is evaluated for numerous bundles of requests. Then the optimal distribution of the bundles among the profit centers is determined via a combinatorial auction. Both approaches propose mechanisms for sharing the resulting profit increase among the centers. Our development was driven by the necessity to control cost effective transportation in an established collaborative network of independent express couriers on the operational level where decisions on the exchange of orders have to be made instantaneously. In our problem environment the situation is highly dynamic such that at no point in time planning of a fixed set of orders/requests is possible as it is assumed in Refs. [10] and [12]. Hence the focus was on implementing rational decision models which allow on-line algorithms within an effective Decision Support System. In this paper we describe our experience with the development and maintenance of pool.tour, a distributed real-time internet-based collaborative Decision Support System (DSS) for a large express courier network, and we analyze the impact of this system on the success factors mentioned above. In a first development we implemented a system proposing order exchanges automatically which are profitable for both partners involved based on the established compensation schema. This technologically highly demanding system has been in use for about 2 years. Yet, the improvement over the formerly used approach where dispatchers had to assume potentials from their experience and to communicate proposals to partners over the phone was much lower than expected. An analysis of the business i.e. the order pool and the proposals generated showed that the system could only rarely find consolidations i.e. insertions of orders in existing trips which were profitable for the acquiring as well as the operating partner but was only able to propose exchanges which result in separate trips. This unsatisfactory behavior motivated us to analyze the contractual compensation schema, to propose an alternative schema and to compare the result with the existing schema. For that purpose we performed a simulation study on the logged order pool over a significant duration of several weeks. Our analysis showed that the rather poor performance could be clearly attributed to the compensation schema. Using a rather straightforward cost based compensation schema within the model base of pool.tour network wide transportation cost could be decreased significantly, even down to the level reachable by centralized planning. Also the distribution of the profit showed to be much fairer. This paper is structured as follows. In Section 2 we give a short description of the transportation market and the specific carrier network underlying our development and study, in the following referred to as the Cooperative Logistic Network (CLN) or simply network for short. In Section 3 we introduce the planning problem and the established compensation schema. Then, in Section 4, we shortly describe our Decision Support System pool.tour. In Section 5 we describe the motivation and the design of our simulation study for evaluating the effectiveness of pool.tour and in Section 6 we report the central results of this experience.

In this paper we have presented an empirical analysis of the effectiveness of a Collaborative Decision Support System in an express carrier network. Due to the highly time sensitive planning environment supporting awareness of consolidation potentials has been identified as a critical success factor and has led to the development and implementation of pool.tour, a technically complex Decision Support System whose core is a real-time proposal generator. The experience of 2 years of operation and the results of our simulation analysis have demonstrated the dominant importance of the choice of an adequate compensation schema. It is another critical success factor for the cooperation network, for the potential of materializing the overall cost reduction as well as for the fairness of profit distribution within the network. Our simulation reveals that with the support of a real-time DSS based on an adequate compensation schema the network is able to work close to the level obtainable by centralized planning. In a sense this is the strongest possible empirical result. Thus, the compensation schema has to be regarded as the primal design parameter of decision support with respect to economic success and stability of the network. It can make or break the network.