No description available.

No description available.

No description available.

We present a hybrid optimization algorithm for mixed-integer linear programming, embedding both heuristic and exact components. In order to validate it we use the periodic vehicle routing problem (PVRP) as a case study. This problem consists of determining a set of minimum cost routes for each day of a given planning horizon, with the constraints that each customer must be visited a required number of times (chosen among a set of valid day combinations), must receive every time the required quantity of product, and that the number of routes per day (each respecting the capacity of the vehicle) does not exceed the total number of available vehicles. This is a generalization of the well-known vehicle routing problem (VRP). Our algorithm is based on the linear programming (LP) relaxation of a set-covering-like integer linear programming formulation of the problem, with additional constraints. The LP-relaxation is solved by column generation, where columns are generated heuristically by an iterated local search algorithm. The whole solution method takes advantage of the LP-solution and applies techniques of fixing and releasing of the columns as a local search, making use of a tabu list to avoid cycling. We show the results of the proposed algorithm on benchmark instances from the literature and compare them to the state-of-the-art algorithms, showing the effectiveness of our approach in producing good quality solutions. In addition, we report the results on realistic instances of the PVRP introduced in Pacheco et al. (2011) [24] and on benchmark instances of the periodic traveling salesman problem (PTSP), showing the efficacy of the proposed algorithm on these as well. Finally, we report the new best known solutions found for all the tested problems. (authors' abstract)

Nowadays traffic congestion is a big problem all over the world. To solve this problem governments build broader roads, establish more reasonable traffic rules. Dynamic routing is a good and efficient way to reduce traffic congestion. That is why research of real time vehicle routing optimization algorithms was made. Experiment showed that best performance of shortest path algorithm was Dijkstra algorithm. Based on that, a software prototype was developed – optimized route search system. Driver can select shortest or fastest route to his destination. There was used roads length to evaluate shortest path and special coefficient to evaluate fastest route. This coefficient is calculated respectively to road load, length, speed limit, capacity. Performance of developed system is good so I conclude that the system is capable of routing vehicles in real time in complex traffic network.

Thesis (Ph. D.)--Lehigh University, 2005. / Includes vita. Includes bibliographical references (leaves 184-192).

Many real-world vehicle routing problems are dynamic optimization problems, with customer requests arriving over time, requiring a repeated reoptimization. In this paper, we consider a dynamic vehicle routing problem where one additional customer arrives at a beforehand unknown location when the vehicles are already under way. Our objective is to maximize the probability that the additional customer can be integrated into one of the otherwise fixed tours without violating time constraints. This is achieved by letting the vehicles wait at suitable locations during their tours, thus influencing the position of the vehicles at the time when the new customer arrives. For the cases of one and two vehicles, we derive theoretical results about the best waiting strategies. The general problem is shown to be NP-complete. Several deterministic waiting strategies and an evolutionary algorithm to optimize the waiting strategy are proposed and compared empirically. It is demonstrated that a proper waiting strategy can significantly increase the probability of being able to service the additional customer, at the same time reducing the average detour to serve that customer.

Interlibrary Loan entails obtaining copies of library materials not found in the library's collection on behalf of the library's patrons (borrowing), as well as providing copies of library materials requested by other libraries (lending). The dynamic nature of today's library environment is well illustrated by the rapid changes occurring in the role of interlibrary loan. The vision statement of the University of Central Florida Library is: The library performs a central role of adding value to information for the academic community by creatively improving and providing information resources and services. The library strives to create an environment that encourages the pursuit of intellectual endeavors and the creation of new knowledge. In an endeavor to fulfill this vision, the Interlibrary Loan Department at the UCF main library wants to set up a document delivery service within the UCF main campus in order to facilitate research efforts on campus. The document delivery service will include delivery and pickup of library materials for ILL requests by faculty online (via computers). In this study, we build a Traveling Salesperson model for obtaining a routing sequence for the document delivery service. Next, we analyze this model in order to check the feasibility of the routing sequence in presence of demand (delivery and pickup) by simulating the demand over the route using computer simulation software. We conclude by validating the model under given conditions and providing route sequence recommendations in the case of extreme demands.

No description available.

In this work we present a multiagent Fleet Coordination Problem (FCP). In this formulation, agents seek to minimize the fuel consumed to complete all deliveries while maintaining acceptable on-time delivery performance. Individual vehicles must both (i) bid on the rights to deliver a load of goods from origin to destination in a distributed, cooperative auction and (ii) choose the rate of travel between customer locations. We create two populations of adaptive agents, each to address one of these necessary functions. By training each agent population in separate source domains, we use transfer learning to boost initial performance in the target FCP. This boost removes the need for 300 generations of agent training in the target FCP, though the source problem computation time was less than the computation time for 5 generations in the FCP. / Graduation date: 2012