Optimal Routing for Shared Autonomous Vehicles Feeder Services in Urban Networks

The Complexities of Operating DAR Services Mean That Computerized Planning and Scheduling Is Necessary for Systems of Realistic Size. This Research Studies an Electric Autonomous Fleet Size with Mix Dial-A-Ride Problem. The Goal of The Problem Is to Minimize a Weighted Objective Function Consisting of The Total Travelling Costs of All Vehicles, Users' Excess Ride Time Costs and Vehicles' Acquisition Costs While Satisfying Customer Service Level Constraints Along with Battery Level Management and Recharge Times Management Constraints. In This Variant of The Dial-A-Ride Problem, Recharging at Any of The Available Charging Stations Is Allowed. A Cluster-First, Route-Second Genetic Algorithm Is Proposed to Solve the Problem, Where the Clustering Is Performed by Choosing the Fleet Size and Assigning the Customers to The Fleet Using a Genetic Algorithm (GA), Then the Primary Routes Are Developed by A Routing Heuristic, Finally the Charging Stations Will Be Inserted to The Algorithm Using an Insertion Technique. The Performance of The Proposed Method Is Tested by Using Benchmark Instances of a Related Problem from The Recent Literature. The Proposed Method Has Achieved Solutions Comparable with The Current State-Of-Art Methods. The Computational Results Show That the Proposed Method Is Effective in Finding Comparable Solutions with The Current State-Of-Art Method. New Instances, Some of Which Include First-Mile Feeder Services, Are Generated Based on The Data from Uber Technologies Inc. Tests Performed on New Instances Demonstrate That the Minimum Possible Fleet Size Does Not Always Result in Minimum Costs. Moreover, The Tests Show That Integration of The Feeder Services into Dial-A-Ride Services Increases Ride-Sharing Ridership.
.