Driven by new laws and regulations concerning the emission of greenhouse gases, carriers are starting to use electric vehicles for last-mile deliveries. The limited battery capacities of these vehicles necessitate visits to recharging stations during delivery tours of industry-typical length, which have to be considered in the route planning to avoid inefficient vehicle routes with long detours. We introduce the electric vehicle-routing problem with time windows and recharging stations E-VRPTW, which incorporates the possibility of recharging at any of the available stations using an appropriate recharging scheme. Furthermore, we consider limited vehicle freight capacities as well as customer time windows, which are the most important constraints in real-world logistics applications. As a solution method, we present a hybrid heuristic that combines a variable neighborhood search algorithm with a tabu search heuristic. Tests performed on newly designed instances for the E-VRPTW as well as on benchmark instances of related problems demonstrate the high performance of the heuristic proposed as well as the positive effect of the hybridization.

The Electric Vehicle-Routing Problem with Time Windows and Recharging Stations

http://hub.hku.hk/bitstream/10722/188944/1/Content.pdf?accept=1

A review of urban transportation network design problems

The Electric Vehicle Routing Problem with Time Windows and Recharging Stations

This paper studies electric vehicle charger location problems and analyzes the impact of public charging infrastructure deployment on increasing electric miles traveled, thus promoting battery electric vehicle (BEV) market penetration. An activity-based assessment method is proposed to evaluate BEV feasibility for the heterogeneous traveling population in the real world driving context. Genetic algorithm is applied to find (sub)optimal locations for siting public charging stations. A case study using the GPS-based travel survey data collected in the greater Seattle metropolitan area shows that electric miles and trips could be significantly increased by installing public chargers at popular destinations, with a reasonable infrastructure investment.

https://lib.dr.iastate.edu/cgi/viewcontent.cgi?article=1131&context=ccee_pubs

Charging infrastructure planning for promoting battery electric vehicles: An activity-based approach using multiday travel data

Multi-horizon forecasting problems often contain a complex mix of inputs -- including static (i.e. time-invariant) covariates, known future inputs, and other exogenous time series that are only observed historically -- without any prior information on how they interact with the target. While several deep learning models have been proposed for multi-step prediction, they typically comprise black-box models which do not account for the full range of inputs present in common scenarios. In this paper, we introduce the Temporal Fusion Transformer (TFT) -- a novel attention-based architecture which combines high-performance multi-horizon forecasting with interpretable insights into temporal dynamics. To learn temporal relationships at different scales, the TFT utilizes recurrent layers for local processing and interpretable self-attention layers for learning long-term dependencies. The TFT also uses specialized components for the judicious selection of relevant features and a series of gating layers to suppress unnecessary components, enabling high performance in a wide range of regimes. On a variety of real-world datasets, we demonstrate significant performance improvements over existing benchmarks, and showcase three practical interpretability use-cases of TFT.

Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecasting

This paper develops an equilibrium modeling framework that captures the interactions among availability of public charging opportunities, prices of electricity, and destination and route choices of plug-in hybrid electric vehicles (PHEVs) at regional transportation and power transmission networks coupled by PHEVs. The modeling framework is then applied to determine an optimal allocation of a given number of public charging stations among metropolitan areas in the region to maximize social welfare associated with the coupled networks. The allocation model is formulated as a mathematical program with complementarity constraints, and is solved by an active-set algorithm. Numerical examples are presented to demonstrate the models and offer insights on the equilibrium of the coupled transportation and power networks, and optimally allocating resource for public charging infrastructure.

Optimal deployment of public charging stations for plug-in hybrid electric vehicles

This paper develops a modeling framework that considers the effect of income on travelers’ choices of trip generation, mode and route on multimodal transportation networks and explicitly captures the distributional impacts of congestion–mitigation policies on different income and geographic groups. The modeling framework is applied to design more equitable yet efficient congestion pricing and tradable credit schemes. The design models are formulated as mathematical programs with equilibrium constraints, and solved by derivative-free solution algorithms. Numerical examples are presented to demonstrate the models and offer insight on the mechanisms of achieving better equity under congestion pricing or tradable credit schemes.

Design of more equitable congestion pricing and tradable credit schemes for multimodal transportation networks

We propose a novel data-driven approach for solving multi-horizon probabilistic forecasting tasks that predicts the full distribution of a time series on future horizons. We illustrate that temporal patterns hidden in historical information play an important role in accurate forecasting of long time series. Traditional methods rely on setting up temporal dependencies manually to explore related patterns in historical data, which is unrealistic in forecasting long-term series on real-world data. Instead, we propose to explicitly learn constructing hidden patterns' representations with deep neural networks and attending to different parts of the history for forecasting the future. In this paper, we propose an end-to-end deep-learning framework for multi-horizon time series forecasting, with temporal attention mechanisms to better capture latent patterns in historical data which are useful in predicting the future. Forecasts of multiple quantiles on multiple future horizons can be generated simultaneously based on the learned latent pattern features. We also propose a multimodal fusion mechanism which is used to combine features from different parts of the history to better represent the future. Experiment results demonstrate our approach achieves state-of-the-art performance on two large-scale forecasting datasets in different domains.

Multi-Horizon Time Series Forecasting with Temporal Attention Learning

Pareto-improving congestion pricing on multimodal transportation networks

This article presents a robust approach for determining optimal locations of public shelters and their capacities, from a given set of potential sites during evacuation planning under demand uncertainty. Demand uncertainty in the article refers to the uncertainty associated with the number of people using the public shelters during evacuation. It is assumed that a planning authority determines the number of shelters, their locations, and capacities whereas evacuees choose a shelter to evacuate and the routes to access it. The proposed model is formulated as a mathematical program with complementarity constraints and is solved by a cutting-plane scheme. A numerical example on the Sioux Falls network demonstrates that robust plans are able to achieve nearly the same level of performance with a significant lower cost as compared to a conservative plan, which assumes the highest demand of each origin node.

Di Wu

Papers

Optimal deployment of public charging stations for plug-in hybrid electric vehicles

Design of more equitable congestion pricing and tradable credit schemes for multimodal transportation networks

Multi-Horizon Time Series Forecasting with Temporal Attention Learning

Pareto-improving congestion pricing on multimodal transportation networks

Robust Shelter Locations for Evacuation Planning with Demand Uncertainty