Center for Logistical Innovations in Freight Systems (CLIFS)

CLIFS Homepage

Approximation Algorithms for Bidding in Combinatorial Auctions for the Procurement of Transportation Service Contracts

Investigator: Amelia C. Regan

Support: U.S.Department of Transportation and California Department of Transportation/ University of California Transportation Center

During the past several years, large shippers have changed the way they contract for trucking services - moving from lane by lane negotiation to using software based loosely on combinatorial auctions. Combinatorial auctions are those in which an auctioneer (shipper) places multiple items out to bid simultaneously and in which the bidders (trucking companies) may present several bids for bundles of items simultaneously. The bid valuation and construction problem for carriers facing combinatorial auctions for the procurement of freight transportation contracts is very difficult and involves the computation of a numbe of NP-hard sub problems. Our research develops computationally tractable approximation methods for estimating values and constructing bids and identifies ways that smaller carriers can work together to receive the benefits available to larger carriers.

The policy implications of this work are not insignificant. In the last twenty year nearly fifty thousand trucking companies went out of business. The health of many US industries is dependent upon the health of the trucking industry and the ability of trucking companies to honor their contracts. In addition, developing tools that will improve the overall efficiency of trucking operations has significant environmental and quality of life impacts for non-commercial vehicle operators.

Door-to-Door Delivery Systems: Operational Changes and New Modeling Challenges

Investigator: Minyoung Park

Support: U.S.Department of Transportation and California Department of Transportation/ University of California Transportation Center

The trucking industry, which plays a crucial role in the economy, is experiencing a transition in both supply and demand as a result of growth advances in information technologies and the growth of e-commerce.

Increasingly, customers of trucking companies including manufacturers, wholesalers, retailers and individual customers, demand quicker and more reliable delivery no matter what the size of the load. The rapidly increasing demand for express package delivery services, a part of the less-than-truckload (LTL) segment in the trucking industry, is noteworthy. These changes on the demand side lead to developments on the supply side. In order to survive in the market, carriers must respond to new customer needs arising from changes in production and distribution systems. A recent survey of trucking firms showed that most traditional trucking firms are investing in advanced information technologies in order to respond to the challenges and opportunities of the new economy in the forms of just-in-time distribution, package express and door-to-door delivery.

Although e-commerce growth has been widely discussed, the impacts of the digital economy on the freight transportation system still remain uncertain, raising many questions and issues such as how and to what extent e-commerce will influence or interact with the freight transportation system and how trucking firms should respond to the changes. In the proposed research, we investigate operational changes arising in the door-to-door express delivery systems and develop an analysis tool incorporating the distinct features of express delivery operations. The results of the research can be used to identify advanced distribution strategies for improving the operating efficiency of door-to-door express delivery systems, while enhancing customer satisfaction.

Dynamic and Stochastic Fleet and Freight Management: Algorithms Development and Performance Analysis

Investigator: Xiangwen Liu

Support: U.S.Department of Transportation and California Department of Transportation/ University of California Transportation Center

Improving the overall efficiency of the freight transportation systems reduces costs to consumers, improves the ability of companies to meet customer service desires and improves national and regional economic strength. Improving the efficiency of emergency and police dispatching saves lives. Improving the efficiency of utility repair fleets saves money and improves the quality of life for affected consumer customers and businesses. The last several years have witnessed a sharp increase in interest in stochastic and dynamic routing and scheduling problems. This dissertation research examines several key problems for stochastic and dynamic routing and scheduling: the probabilistic traveling salesman problem, the dynamic traveling salesman problem and the dynamic traveling repairman problem. We combine stochastic optimization, queuing theory analysis and static optimization to develop and analyze algorithms for these problems. Our preliminary results extend results provided by other logistics researchers and computer scientists. Namely that if we apply optimal deterministic static solutions properly in a dynamic and stochastic environment, that we can obtain the asymptotically optimal solution for the dynamic problem.

E-commerce and the efficiency of the California freight network: Perspectives of shippers, carriers and third party logistics and information services providers

Investigators: Thomas F. Golob and Amelia C. Regan

Support: U.S.Department of Transportation and California Department of Transportation/ University of California Transportation Center

A substantial portion of all business-to-business transactions in the U.S. now occurs online or electronically, over private specialized networks using EDI (electronic data interchange) and over the Internet. Such business-to-business "e-commerce" is expected to increase dramatically, particularly the share that occurs on the Internet. Online activities include, but are not limited to, issuing catalogs, quotes and schedules, placing orders and bids, and consolidating, scheduling, and tracking of shipments. While e-commerce is increasing productivity, it is not known how it is affecting freight transportation, and particularly urban goods movement. We propose to study the impact of e-commerce and related information technology on shippers, carriers, and third party logistics and information providers by analyzing data from three new closely related and simultaneously implemented Internet-based (online) industry surveys. The Internet surveys will be used to gather industry representatives' perceptions of e- commerce impacts on freight demand by mode and by the size and timing of shipments. The surveys will also explore which aspects of e-commerce can be most effective in alleviating congestion on the freight network. These data will be analyzed using discrete choice models and multivariate statistical methods to describe technology adoption and to allow forecasting of e-commerce effects on freight demand patterns.

Policies for Safer Truck Operations on Urban Freeways

Investigators: Thomas F. Golob and Amelia C. Regan

Support: U.S.Department of Transportation and California Department of Transportation/ University of California Transportation Center

Trucks carry 63% of all freight shipments in by ton-mile and 72% of all shipments by value. The shippers and receivers of most of this trucked freight, as well as the major intermodal facilities (particularly seaports and airports) are located within California's major urban areas. Consequently, freeways with the heaviest truck traffic are also those that carry heavy volumes of commuters. Customer schedules for pickup and delivery and intermodal operations require that trucks operate on these freeways during congested periods. The California Department of Transportation monitors truck traffic volumes and also maintains a database of traffic accidents. Preliminary analyses of these crash data in combination with truck volume data factored by time of day show that truck-involved crashes are more likely on certain freeway segments, after controlling for differences in levels of truck exposure. We propose to use appropriate statistical analysis methods to identify freeway locations and times at which trucks mixes are problematic. We then propose to survey the literature and consult both trucking industry and metropolitan planning sources to identify and evaluate policy initiatives that can be used to re-route and re-schedule some truck traffic to safer places and times.