Exact and meta-heuristic approach for a general heterogeneous dial-a-ride problem with multiple depots

Dial-a-ride problems are concerned with the design of efficient vehicle routes for transporting individual persons from specific origin to specific destination locations. In real-life this operational planning problem is often complicated by several factors. Users may have special requirements (e.g. to be transported in a wheelchair) while service providers operate a heterogeneous fleet of vehicles from multiple depots in their service area. In this paper, a general dial-a-ride problem in which these three real-life aspects may simultaneously be taken into account is introduced: the Multi-Depot Heterogeneous Dial-A-Ride Problem (MD-H-DARP). Both a three- and two-index formulation are discussed. A branch-and-cut algorithm for the standard dial-a-ride problem is adapted to exactly solve small problem instances of the MD-H-DARP. To be able to solve larger problem instances, a new deterministic annealing meta-heuristic is proposed. Extensive numerical experiments are presented on different sets of benchmark instances for the homogeneous and the heterogeneous single depot dial-a-ride problem. Instances for the MD-H-DARP are introduced as well. The branch-and-cut algorithm provides considerably better results than an existing algorithm which uses a less compact formulation. All seven previously unsolved benchmark instances for the heterogeneous dial-a-ride problem could be solved to optimality within a matter of seconds. While computation times of the exact algorithm increase drastically with problem size, the proposed meta-heuristic algorithm provides near-optimal solutions within limited computation time for all instances. Several best known solutions for unsolved instances are improved and the algorithm clearly outperforms current state-of-the-art heuristics for the homogeneous and heterogeneous dial-a-ride problem, both in terms of solution quality and computation time.     

A distributed VNS algorithm for optimizing dial-a-ride problems in large-scale scenarios

These days, transportation and logistic problems in large cities are demanding smarter transportation services that provide flexibility and adaptability. A possible solution to this arising problem is to compute the best routes for each new scenario. In this problem, known in the literature as the dial-a-ride problem, a number of passengers are transported between pickup and delivery locations trying to minimize the routing costs while respecting a set of prespecified constraints. This problem has been solved in the literature with several approaches from small to medium sized problems. However, few efforts have dealt with large scale problems very common in massive scenarios (big cities or highly-populated regions). In this study, a new distributed algorithm based on the partition of the requests space and the combination of the routes is presented and tested on a set of 24 different scenarios of a large-scale problem (up to 16,000 requests or 32,000 locations) in the city of San Francisco. The results show that, not only the distributed algorithm is able to solve large problem instances that the corresponding sequential algorithm is unable to solve in a reasonable time, but also to have an average improvement of 9% in the smaller problems. The results have been validated by means of statistical procedures proving that the distributed algorithm can be an effective way to solve high dimensional dial-a-ride problems.     

Multiple objective optimization of the fleet sizing problem for road freight transportation

A fleet sizing problem (FSP) in a road freight transportation company with heterogeneous fleet and its own technical back‐up facilities is considered in the paper. The mathematical model of the decision problem is formulated in terms of multiple objective mathematical programming based on queuing theory. Technical and economical criteria as well as interests of different stakeholders are taken into account in the problem formulation. The solution procedure is composed of two steps. In the first one a sample of Pareto‐optimal solutions is generated by an original program called MEGROS. In the second step this set is reviewed and evaluated, according to the Decision Maker's (DM's) model of preferences. The evaluation of solutions is carried out with an application of an interactive multiple criteria analysis method, called Light Beam Search (LBS). Finally, the DM selects the most desirable, compromise solution.     

Solving the Fleet Size and Mix Vehicle Routing Problem with Time Windows via Adaptive Memory Programming

This paper presents a novel Adaptive Memory Programming (AMP) solution approach for the Fleet Size and Mix Vehicle Routing Problem with Time Windows (FSMVRPTW). The FSMVRPTW seeks to design a set of depot returning vehicle routes to service a set of customers with known demands, for a heterogeneous fleet of vehicles with different capacities and fixed costs. Each customer is serviced only once by exactly one vehicle, within fixed time intervals that represent the earliest and latest times during the day that service can take place. The objective is to minimize the total transportation costs, or similarly to determine the optimal fleet composition and dimension following least cost vehicle routes. The proposed method utilizes the basic concept of an AMP solution framework equipped with a probabilistic semi-parallel construction heuristic, a novel solution re-construction mechanism, an innovative Iterated Tabu Search algorithm tuned for intensification local search and frequency-based long term memory structures. Computational experiments on well-known benchmark data sets illustrate the efficiency and effectiveness of the proposed method. Compared to the current state-of-the-art, the proposed method improves the best reported cumulative and mean results over most problem instances with reasonable computational requirements.     

带时间窗的混和车辆路径问题的优化算法研究

文章针对带时间窗约束的混合车辆路径问题的特点,建立了带时间窗的混合车辆路径问题的数学模型,并设计了变邻域禁忌搜索算法对该问题进行求解。通过标准算例测试及与现有文献计算结果的比较,验证了该算法的有效性。     

Congestion-aware system optimal route choice for shared autonomous vehicles

We study the shared autonomous vehicle (SAV) routing problem while considering congestion. SAVs essentially provide a dial-a-ride service to travelers, but the large number of vehicles involved (tens of thousands of SAVs to replace personal vehicles) results in SAV routing causing significant congestion. We combine the dial-a-ride service constraints with the linear program for system optimal dynamic traffic assignment, resulting in a congestion-aware formulation of the SAV routing problem. Traffic flow is modeled through the link transmission model, an approximate solution to the kinematic wave theory of traffic flow. SAVs interact with travelers at origins and destinations. Due to the large number of vehicles involved, we use a continuous approximation of flow to formulate a linear program. Optimal solutions demonstrate that peak hour demand is likely to have greater waiting and in-vehicle travel times than off-peak demand due to congestion. SAV travel times were only slightly greater than system optimal personal vehicle route choice. In addition, solutions can determine the optimal fleet size to minimize congestion or maximize service.     

Transit network design with allocation of green vehicles: A genetic algorithm approach

The use of fossil fuels in transportation generates harmful emissions that accounts for nearly half of the total pollutants in urban areas. Dealing with this issue, local authorities are dedicating specific efforts to seize the opportunity offered by new fuels and technological innovations in achieving a cleaner urban mobility. In fact, authorities are improving environmental performances of their public transport fleet by procuring cleaner vehicles, usually called low and zero emission vehicles (LEV and ZEV, respectively). Nevertheless there seems to be a lack of methodologies for supporting stakeholders in decisions related to the introduction of green vehicles, whose allocation should be performed since the network design process in order to optimize their available green capacity.In this paper, the problem of clean vehicle allocation in an existing public fleet is faced by introducing a method for solving the transit network design problem in a multimodal, demand elastic urban context dealing with the impacts deriving from transportation emissions.The solving procedure consists of a set of heuristics which includes a routine for route generation and a genetic algorithm for finding a sub-optimal set of routes with the associated frequencies.     

A multi-period dial-a-ride problem with driver consistency

Dial-a-ride services provide disabled and elderly people with a personalized mode of transportation to preserve their mobility. Typically, several users with different pickup and dropoff locations are transported on a vehicle simultaneously. The focus in dial-a-ride problems (DARPs) is mainly on minimizing routing cost. Service quality has been taken into account in the models by imposing time windows and limiting the maximum ride time of each user. We extend the classical DARP by an additional feature of service quality referred to as driver consistency. Customers of dial-a-ride services are often sensitive to changes in their daily routine. This aspect includes the person who is providing the transportation service, i.e., the driver of the vehicle. Our problem, called the driver consistent dial-a-ride problem (DC-DARP), considers driver consistency by bounding the maximum number of different drivers that transport a user over a multi-period planning horizon.We propose different formulations of the problem and examine their efficiency when applied in a Branch-and-Cut fashion. Additionally, we develop a large neighborhood search algorithm that generates near-optimal solutions in a short amount of time.Over 1000 instances are generated with close reference to real world scenarios. Extensive computational experiments are conducted in order to assess the quality of the solution approaches and to provide insights into the new problem. Results reveal that the cost of offering driver consistency varies greatly in magnitude. Depending on the instance, the cost of assigning one driver to each user can be up to 27.98% higher compared to a low-cost solution. However, routing cost increases by not more than 5.80% if users are transported by at least two drivers.     

Just-in-time delivery for green fleets: A feedback control approach

With increasing attention being paid to greenhouse gas (GHG) emissions, the transportation industry has become an important focus of approaches to reduce GHG emissions, especially carbon dioxide equivalent (CO₂e) emissions. In this competitive industry, of course, any new emissions reduction technique must be economically attractive and contribute to good operational performance. In this paper, a continuous-variable feedback control algorithm called GEET (Greening via Energy and Emissions in Transportation) is developed; customer deliveries are assigned to a fleet of vehicles with the objective function of Just-in-Time (JIT) delivery and fuel performance metrics akin to the vehicle routing problem with soft time windows (VRPSTW). GEET simultaneously determines vehicle routing and sets cruising speeds that can be either fixed for the entire trip or varied dynamically based on anticipated performance. Dynamic models for controlling vehicle cruising speed and departure times are proposed, and the impact of cruising speed on JIT performance and fuel performance are evaluated. Allowing GEET to vary cruising speed is found to produce an average of 12.0–16.0% better performance in fuel cost, and −36.0% to +16.0% discrepancy in the overall transportation cost as compared to the Adaptive Large Neighborhood Search (ALNS) heuristic for a set of benchmark problems. GEET offers the advantage of extremely fast computational times, which is a substantial strength, especially in a dynamic transportation environment.     

Jointly optimizing cost, service, and environmental performance in demand-responsive transit scheduling

In certain fleet systems, the environmental impacts of operation are, to some extent, a controllable function of vehicle routing and scheduling decisions. However, little prior work has considered environmental impacts in fleet vehicle routing and scheduling optimization, in particular, where the impacts were assessed systematically utilizing life-cycle impact assessment methodologies such as those described by the Society of Environmental Chemistry and Toxicology. Here a methodology is presented for the joint optimization of cost, service, and life-cycle environmental consequences in vehicle routing and scheduling, which we develop for a demand-responsive (paratransit or dial-a-ride) transit system. We demonstrate through simulation that, as a result of our methodology, it is possible to reduce environmental impacts substantially, while increasing operating costs and service delays only slightly.     

The fleet size and mix pollution-routing problem

This paper introduces the fleet size and mix pollution-routing problem which extends the pollution-routing problem by considering a heterogeneous vehicle fleet. The main objective is to minimize the sum of vehicle fixed costs and routing cost, where the latter can be defined with respect to the cost of fuel and CO₂ emissions, and driver cost. Solving this problem poses several methodological challenges. To this end, we have developed a powerful metaheuristic which was successfully applied to a large pool of realistic benchmark instances. Several analyses were conducted to shed light on the trade-offs between various performance indicators, including capacity utilization, fuel and emissions and costs pertaining to vehicle acquisition, fuel consumption and drivers. The analyses also quantify the benefits of using a heterogeneous fleet over a homogeneous one.     

Multiple criteria optimization method for the vehicle assignment problem in a bus transportation company

A vehicle assignment problem (VAP) in a road, long‐haul, passenger transportation company with heterogeneous fleet of buses is considered in the paper. The mathematical model of the VAP is formulated in terms of multiobjective, combinatorial optimization. It has a strategic, long‐term character and takes into account four criteria that represent interests of both passengers and the company's management. The decision consists in the definition of weekly operating frequency (number of rides per week) of buses on international routes between Polish and Western European cities. The VAP is solved in a step‐wise procedure. In the first step a sample of efficient (Pareto‐optimal) solutions is generated using an original metaheuristic method called Pareto Memetic Algorithm (PMA). In the second step this sample is reviewed and evaluated by the Decision Maker (DM). In this phase an interactive, multiple criteria analysis method with graphical facilities, called Light Beam Search (LBS), is applied. The method helps the DM to define his/her preferences, direct the search process and select the most satisfactory solution.     

Design framework of large-scale one-way electric vehicle sharing systems: A continuum approximation model

This paper proposes a Continuum Approximation (CA) model for design of a one-way Electrical Vehicle (EV) sharing system that serves a metropolitan area. This model determines the optimal EV sharing station locations and the corresponding EV fleet sizes to minimize the comprehensive system cost, including station construction investment, vehicle charging, transportation and vehicle balancing, under stochastic and dynamic trip demands. This is a very complex problem due to the NP-hard nature of location design, the large number of individual users, and the stochasticity and dynamics of generated trips. Further, the considerable charging time required by EVs distinguishes this problem from traditional car sharing problems where a vehicle is immediately available for pickup after being dropped at a station. We find that the CA approach can overcome these modeling challenges by decomposing the studied area into a number of small neighborhoods that each can be approximated by an Infinite Homogeneous Plane (IHP). We find that the system cost of an IHP is a unimodal function of the station service area size and can be efficiently solved in a sub-linear time by the bisection algorithm. Then integrating the solutions of all IHPs yields an approximate solution to the original heterogeneous area. With numerical experiments, we show that the CA solution is able to estimate the total system cost of the discrete counterpart solution efficiently with good accuracy, even for large-scale heterogeneous problems. This implies that the proposed CA approach is capable of providing a near-optimum solution to the comprehensive design of a practical large-scale EV sharing system. With this model, we also conduct sensitivity analysis to reveal insights into how cost components and system design vary with key parameter values. As far as the author’s knowledge, this study is the first work that addresses design of an EV sharing system considering both longer-term location and fleet size planning and daily vehicle operations. The proposed CA model also extends the CA methodology literature from traditional location problems with stationary demand, single-facility based service to EV sharing problems considering dynamic demands, OD trips, and nonlinear vehicle charging times.     

Energy impact evaluation for eco-routing and charging of autonomous electric vehicle fleet: Ambient temperature consideration

This paper studies the heterogeneous energy cost and charging demand impact of autonomous electric vehicle (EV) fleet under different ambient temperature. A data-driven method is introduced to formulate a two-dimensional grid stochastic energy consumption model for electric vehicles. The energy consumption model aids in analyzing EV energy cost and describing uncertainties under variable average vehicle trip speed and ambient temperature conditions. An integrated eco-routing and optimal charging decision making framework is designed to improve the capability of autonomous EV’s trip level energy management in a shared fleet. The decision making process helps to find minimum energy cost routes with consideration of charging strategies and travel time requirements. By taking advantage of derived models and technologies, comprehensive case studies are performed on a data-driven simulated transportation network in New York City. Detailed results show us the heterogeneous energy impact and charging demand under different ambient temperature. By giving the same travel demand and charging station information, under the low and high ambient temperature within each month, there exist more than 20% difference of overall energy cost and 60% difference of charging demand. All studies will help to construct sustainable infrastructure for autonomous EV fleet trip level energy management in real world applications.     

Siting charging stations for electric vehicle adoption in shared autonomous fleets

The transportation sector is undergoing three revolutions: shared mobility, autonomous driving, and electrification. When planning the charging infrastructure for electric vehicles, it is critical to consider the potential interactions and synergies among these three emerging systems. This study proposes a framework to optimize charging infrastructure development for increasing electric vehicle (EV) adoption in systems with different levels of autonomous vehicle adoption and ride sharing participation. The proposed model also accounts for the pre-existing charging infrastructure, vehicle queuing at the charging stations, and the trade-offs between building new charging stations and expanding existing ones with more charging ports.Using New York City (NYC) taxis as a case study, we evaluated the optimum charging station configurations for three EV adoption pathways. The pathways include EV adoption in a 1) traditional fleet (non-autonomous vehicles without ride sharing), 2) future fleet (fully autonomous vehicles with ride sharing), and 3) switch-over from traditional to future fleet. Our results show that, EV adoption in a traditional fleet requires charging infrastructure with fewer stations that each has more charging ports, compared to the future fleet which benefits from having more scattered charging stations. Charging will only reduce the service level by 2% for a future fleet with 100% EV adoption. EV adoption can reduce CO₂ emissions of NYC taxis by up to 861 Tones/day for the future fleet and 1100 Tones/day for the traditional fleet.     

Integrated scheduling of m-truck,m-drone,and m-depot constrained by time-window,drop-pickup,and m-visit using constraint programming

The idea of deploying unmanned aerial vehicles, also known as drones, for final-mile delivery in logistics operations has vitalized this new research stream. One conceivable scenario of using a drone in conjunction with a traditional delivery truck to distribute parcels is discussed in earlier literature and termed the parallel drone scheduling traveling salesman problem (PDSTSP). This study extends the problem by considering two different types of drone tasks: drop and pickup. After a drone completes a drop, the drone can either fly back to depot to deliver the next parcels or fly directly to another customer for pickup. Integrated scheduling of multiple depots hosting a fleet of trucks and a fleet of drones is further studied to achieve an operational excellence. A vehicle that travels near the boundary of the coverage area might be more effective to serve customers that belong to the neighboring depot. This problem is uniquely modeled as an unrelated parallel machine scheduling with sequence dependent setup, precedence-relationship, and reentrant, which gives us a framework to effectively consider those operational challenges. A constraint programming approach is proposed and tested with problem instances of m-truck, m-drone, m-depot, and hundred-customer distributed across an 8-mile square region.     

Recursive expected conditional value at risk in the fleet renewal problem with alternative fuel vehicles

We study the fleet portfolio management problem faced by a firm deciding which alternative fuel vehicles (AFVs) to choose for its fleet to minimise the weighted average of cost and risk, in a stochastic multi-period setting. We consider different types of technology and vehicles with heterogeneous capabilities. We propose a new time consistent recursive risk measure, the Recursive Expected Conditional Value at Risk (RECVaR), which we prove to be coherent. We then solve the problem for a large UK based company, reporting how the optimal policies are affected by risk aversion and by the clustering for each type of vehicle.     

An analysis on long term emission benefits of a government vehicle fleet replacement plan in northern illinois

There have been a number of studies of the effectiveness of vehicle scrappage programs, which offer incentives to accelerated scrappage of older vehicles often thought to be high emitters. These programs are voluntary and aimed at replacement of household vehicles. In contrast, there is a gap in knowledge related to the emissions benefits of government fleet replacement (retirement) programs. In this study, the efficacy of a fleet replacement program for a local government agency in Northern Illinois, the Forest Preserve of DuPage County (FPDC), is examined using a probabilistic vehicle survival model that accounts for time-varying covariates such as vehicle age and gasoline price. The vehicle lifetime operating emissions are calculated based on the estimated vehicle survival probabilities from the survival model and compared with those derived using the EPA default fleet used in MOBILE6 and the fleet represented by the Oak Ridge National Laboratory (ORNL) survival curve. The results suggest that while there may be short term emission benefits of the FPDC fleet replacement plan, the long-term emission benefits are highly sensitive to economic factors (e.g., future gasoline price) and exhibit a decreasing trend. This indicates that an adaptive multi-stage replacement strategy as opposed to a fixed one is preferable to achieve optimal cost effectiveness.

A new statistical method of assigning vehicles to delivery areas for CO2 emissions reduction

Transportation CO₂ emissions are expected to increase in the following decades, and thus, new and better alternatives to reduce emissions are needed. Road transport emissions are explained by different factors, such as the type of vehicle, delivery operation and driving style. Because different cities may have conditions that are characterized by diversity in landforms, congestion, driving styles, etc., the importance of assigning the proper vehicle to serve a particular region within the city provides alternatives to reduce CO₂ emissions. In this article, we propose a new methodology that results in assigning trucks to deliver in areas such that the CO₂ emissions are minimized. Our methodology clusters the delivery areas based on the performance of the vehicle fleet by using the k-means algorithm and Tukey’s method. The output is then used to define the optimal CO₂ truck-area assignment. We illustrate the proposed approach for a parcel company that operates in Mexico City and demonstrate that it is a practical alternative to reduce transportation CO₂ emissions by matching vehicle type with delivery areas.     

Development of a microscopic activity-based framework for analyzing the potential impacts of transportation control measures on vehicle emissions

The 1990 Clean Air Act Amendments (CAAA) and the Intermodal Surface Transportation Efficiency Act of 1991 (ISTEA) have defined a set of transportation control measures to counter the increase in the vehicle emissions and energy consumption due to increased travel. The value of these TCM strategies is unknown as there is limited data available to measure the travel effects of individual TCM strategies and the models are inadequate in forecasting changes in travel behavior resulting from these strategies. The work described in this paper begins to provide an operational methodology to overcome these difficulties so that the impacts of the policy mandates of both CAAA and ISTEA can be assessed. Although the framework, as currently developed, falls well short of actually forecasting changes in traveler behavior relative to policy options designed to encourage emissions reduction, the approach can be useful in estimating upper bounds of certain policy alternatives in reducing vehicle emissions. Subject to this important limitation, the potential of transportation policy options to alleviate vehicle emissions is examined in a comprehensive activity-based approach. Conclusions are drawn relative to the potential emissions savings that can be expected from efficient trip chaining behavior, ridesharing among household members, as well as from technological advances in vehicle emissions control devices represented by replacing all of the vehicles in the fleet by vehicles conforming to present-day emissions technology.