1、 1 外文翻译 原文 Pareto-Improving Contracts for Express Package Delivery Services Material Source: University of California Author: Alexandra M. Newman Abstract： We address the problem of an express package delivery company in structuring a long-term customer contract whose terms may include prices that d
2、iffer by day-of-week and by speed-of-service. The company traditionally offered speed-of-service pricing to its customers, but without day-of-week differentiation,resulting in customer demands with considerable day-of-week seasonality. The package delivery company hoped that using day-of-week and sp
3、eed-of-service price differentiation for contract customers would induce these customers to adjust their demands to become counter-cyclical to the non-contract demand. Although this usually cannot be achieved by pricing alone, we devise an approach that utilizes day-of-week and speed-of-service pric
4、ing as an element of a Pareto-improving contract. The contract provides the lowest-cost arrangement for the package delivery company while ensuring that the customer is at least as well off as he would have been under the existing pricing structure. The contract pricing smoothes the package delivery
5、 companys demand and reduces peak requirements for transport capacity. The latter helps to decrease capital costs, which may allow a further price reduction for the customer. We formulate the pricing problem as a biconvex optimization model, and present a methodology for designing the contract and n
6、umerical examples that illustrate the achievable savings. Keywords： transportation contracts; contract pricing; speed-of-service pricing; time-of-use pricing; day-of-week pricing I. INTRODUCTION Most package delivery companies (PDCs) charge a premium for faster delivery, but the practice of pricing
7、by day of week is very limited. In the absence of this type of price differentiation, shipment volumes exhibit strong day-of-week patterns, especially in the express package delivery market. Although the schedules of various ground transport vehicles often can be adjusted to account for this day-of-
8、week seasonality, express package delivery companies rely heavily on aircraft, for which it is not possible to match shipping capacity to demand very well. Consequently, excess shipping capacity varies by day of week and by route.When negotiating with potential high-volume contract customers, it may
9、 be advantageous to offer the customer an 2 incentive to release packages countercyclically to the overall demand pattern. Such a counter-cyclical release pattern would improve the profit of the PDC in two ways. First, revenue is generated using available excess capacity for which the incremental op
10、erating costs are quite small. Second, by smoothing the overall demand pattern, requirements for additional transport capacity (typically provided by commercial carriers at premium prices) are minimized, and the PDC is able to provide more reliable service to all customers because the reduced peak l
11、oads pose less strain on pickup, delivery, and sortation resources. Because the incremental cost of servicing a contract customer with a counter-cyclical demand pattern may be small, the PDC may be able to pass on the savings to its customers by charging lower average prices. Our research was motiva
12、ted by a PDC whose management had hoped to induce the companys contract customers to behave in the desired way via day-ofweek and speed-of-service pricing alone. As we explain in more detail later, this is usually not possible. For this reason, we seek to develop a methodology for structuring contra
13、ctswhich may include day-ofweek and speed-of-service pricing as one element that achieves the highest total profit for the PDC while ensuring that the customer is at least as well off as he would be under an existing contract or under any arbitrary reference price structure. We examine this problem
14、assuming that the PDC is negotiating with one major customer at a time. The most promising opportunities for improving the PDCs profit via more complex contract pricing arrangements occur in situations in which several customers sharing an aircraft route have similar day-ofweek seasonality. This phe
15、nomenon occurs frequently due to weekly procurement cycles. For example, automobile assembly plants request deliveries of many parts on Monday morning to supply the assembly line for the week. (Although this may not be optimal,typical material requirements planning systems operate on a weekly schedu
16、le, and the procurement process follows suit.) Component suppliers in the same vicinity that provide parts to a given assembly plant therefore ship on the same day, usually Friday. The PDC would like all of these customers to modify their shipment plans, but it usually faces the problem of negotiati
17、ng with them one at a time. When negotiating with a given customer, the PDC could consider likely outcomes of later negotiations with other customers, but this is obviously difficult to do because of the uncertainty involved. In our approach, various problem data can be specified to account for any
18、capacity availability profiles (induced by non-contract customers and other contract 3 customers) that the PDC wishes to consider. In this paper, we focus on the flow of a class of homogeneous (or nearly homogeneous) packages from a single shipper (typically a manufacturer) that provides vendor-mana
19、ged inventory (VMI) services to a single consignee (a downstream user of the manufactured parts). In the concluding section, we explain how our approach can be generalized to multiple package types. Because of the VMI arrangement, the shipper owns the goods and therefore incurs inventory holding cos
20、ts until the consignee utilizes the goods. We emphasize that our approach is designed for situations in which the customer has considerable control over the timing of package releases which would usually entail changes in the production schedule, and thus our approach probably would not be suitable
21、for an Internet retailer that is expected to fulfill orders soon after they arrive, often by a speed or mode of service chosen by the end-customer. The remainder of this paper is organized as follows: The next section contains a review of the literature. This is followed by formal statements of the
22、PDCs and customers decision problems. In Section 4, we formulate the PDCs and customers problems under a price-only contract and discuss the shortcomings of such a contract in our problem context, and this discussion provides a backdrop for our solution strategy. In Section 5, we present the details
23、 of our methodology for structuring Pareto-improving contracts. Section 6 provides numerical examples that illustrate our proposed method and its benefits. Section 7 closes the paper with a discussion of extensions and generalizations of our approach. 2. Literature Review In this section, we provide
24、 an overview of the separate literatures on time-of-service pricing, and on speedof-service and priority pricing. It is important to point out that, to the best of our knowledge, there is very little research that considers both simultaneously. We first discuss time-of-service pricing with an emphas
25、is on electricity, toll roads, and computer and telecommunication network services, which are the most common application areas. Later in the section, we discuss the literature on speed-of-service and priority pricing, which tends to be less application-specific. In the interest of brevity, our cita
26、tions are limited. Our intent is to provide the reader a sense of the issues that have been explored. 2.1. Time-of-Service Pricing 4 Vickrey (1971) provides a very lucid qualitative discussion of the benefits of what he calls “responsive pricing,” that is, pricing that varies according to the state
27、of the system. Responsive pricing includes such concepts as dynamic pricing based on instantaneous (real-time) congestion, time-of-service pricing based on forecasted (not real-time) demand or congestion patterns, and pricing schemes in the vein of currentday revenue management. Vickrey (1971) menti
28、ons application areas such as long-distance telephone service, airline reservations, and water and power deliverythe very same types of applications that motivate present-day research. 2.1.1. Electricity. Electricity markets are the most common application domain for time-of-use pricing, which is co
29、mmonly referred to as peak load pricing in this industry. Here, peak prices have the effect of both reducing total demand and shifting some demand to off-peak periods. Most of the research can be classified into three broad areas: (1) the welfare economics of time-of-use pricing, (2) models of price
30、 elasticity for electricity, and (3) methods for setting prices. Seminal papers on the welfare benefits of peak-load pricing include Boiteux (1960) and Williamson (1966). Although much of the discussion is posed in terms of peak versus off-peak prices, Panzar (1976) argues that capacity costs depend
31、 not only on peak loads but also on the loads during non-peak periods. Eckel (1987) examines the question of pricing based on demandclass (i.e., industrial, commercial, and residential consumers). The literature on models of price elasticity for electricity is too extensive to discuss here. For a re
32、cent article, see Kamerschen and Porter (2004). These price elasticity models and estimates are widely used in pricing methods, where the emphasis is on setting prices during peak demand periods so as to attenuate demand and thereby reduce capacity requirements. Crew et al. (1995) provide a historic
33、al perspective on optimization-based time-of-use pricing approaches, focusing on non-storable goods such as electricity. Borenstein (2005) highlights several important issues in designing a pricing scheme for utility companies, including: (1) how often prices change, and (2) how long the delay betwe
34、en setting and realizing a price is. The most extreme, yet most effective, pricing scheme is real-time pricing. Borenstein describes various implementations of real-time pricing, and the implications of each. He notes that technology plays a key role in the effectiveness of real-time pricing. 2.1.2.
35、 Transportation. Although peak pricing is not yet widespread in transportation systems, 5 researchers have been espousing the welfare gains and social benefits for years, citing the need to consider factors such as congestion externalities and environmental effects. See, e.g., an early paper by Vick
36、rey (1963) and a more recent anthology edited by Button and Verhoef (1998). Wachs (2005) describes the current state of peak-load pricing on urban road networks, noting that only recently has technology enabled such pricing methods. More recent research on time-of-day pricing for toll roads, bridges
37、, tunnels, etc., has begun to consider the impact of traveler choices. Generally, these models assume that the traveler has the objective of minimizing some function of delay and out-of-pocket costs, and the toll setter chooses prices to maximize social welfare. Examples of papers in this stream of
38、research include Arnott et al. (1990), Yan and Lam (1996), and Daganzo and Garcia (2000). Two papers that treat models similar to ours are Brotcorne et al. (2000 and 2001). The authors address static problems in which the transport provider sets day-of-week (but not speed-of-service) prices and the
39、customer chooses how much to ship on each day to satisfy some aggregate requirement over the horizon. An important simplifying assumption in these models is that the PDC has infinite capacity to handle each of the customers shipping options. 2.1.3. Computer Network and Telecommunication Services. Co
40、mputer network and Internet services represent another important application arena for peak-load pricing because of the very high amplitude of peaks that cause “busy signals” and slow transmission. At this writing, time-of-use pricing is rarely used, and many vendors promote flat rate, rather than u
41、sage-based, pricing. Researchers have modeled and demonstrated the benefits of pricing based both on usage alone and on usage in combination with induced congestion externalities (Gupta et al. 2001) for computer networks. For both computer network and Internet services, Paschalidis and Tsitsiklis (2
42、000) suggest that time-of-day pricing alone, without adjustments for instantaneous congestion, may be sufficient to achieve good results for revenue and welfare maximization. In their model, customers are classified by average processing time per “call,” and each customer class pays a different fee.
43、 In the context of pricing communications bandwidth, Altmann and Chu (2001) propose a combination of a flat fee that covers the cost of a basic level of bandwidth and usage-based charges for on-demand access to higher levels of bandwidth. Interestingly, telecommunications service providers have long
44、 used time-of-day 6 and day-of-week pricing, but as telecommunications capacity expands and competition becomes fiercer, vendors are offering more flat-rate, unlimited-use packages. These patterns are consistent with observations by Odlyzko (2001) who reports that for various communication technolog
45、ies from regular mail to the Internet, as the technology matures, quality improves, prices fall, and pricing plans become simpler. 2.1.4. Time-of-Service Pricing in Other Industries. Increasingly more sophisticated time-of-service pricingoften called “revenue management” in recent yearshas been adop
46、ted in industries in which many customers make purchases or reservations in advance. Gerstner (1986) examines peak-load pricing for private enterprises such as airlines, hotels, and restaurants. These scenarios differ from most of those above because of the need to consider competition, either direc
47、tly or indirectly. For surveys, see Weatherford and Bodily (1992), Bitran and Caldentey (2003), and Talluri and van Ryzin (2004). We now turn to a discussion of pricing based on speed of service or service priority. 译文 帕雷托 改进式 包裹递送服务合同 资料来源 : 加利福尼亚大学 作者： 亚历山德拉 .纽曼 摘要 : 我们要解决快递公司在建立长期客户合约上的问题，该合约的条款可
48、能包括因 日 周 和速度 服务而不同的定价。快递公司传统上为客户提供速度 服务价格，而没有日 周的区别，这导致了客户要求日 周的季节性差价。快递公司希望在合同中运用日 周和速度 服务定价差异，客户会促使他们的客户调节他们的需求以 能够对合约外的需求进行反向循环。尽管这通常不能够仅通过定价来实现，但是我们设计了一种方法，利用日 周和速度 服务的定价方式作为帕累托改进合同的一个要素。该合同为快递公司提供代价最低的安排，而能保证客户至少能够得到其在现有定价结构下的好处。该合同的定价方式满足了快递公司的需求，也降低了运输能力的最高要求。后者能够帮助降低资本话费，这可能会使客户的价格也相应降低。我们将定
49、价问题作为两面都是最优化的模型进行系统阐述，并且也展示设计合同的方法，以及能够说明可实现节余的无数个例子。 关键词：运输合同；合同定价；速度 服务定价； 实时定价；日 周定价 7 引言 大多数快递公司对加快的邮件要求额外补贴，但是实际上按周算的天数的定价是有限的。缺少了这种类型的定价差异，船运总量显示出了强烈的日 周形式，尤其在快递市场上。虽然各种基础运输车辆的日程可以根据日 周的季节性差异进行调节，但是快递公司严重依赖于飞机，因为要求装货容量很好地达到需求是不可能的。结果，额外的装货容量在日 周和路径上有所有不同。当与可能是高额合同的客户谈判时，为客户提供减缓包裹反向循环的动机在整个需求模式中是有利的。这样的反向循环减缓模式将会在两方面提高快递公司 的利润。第一，使用额外容量能产生收益，因为增加的运作费用非常少。第二，运用整个需求模式，额外的运输能力要求（通常由商务运输公
