The Probabilistic Travelling Salesman Problem with Crowdsourcing

We study a variant of the Probabilistic Travelling Salesman Problem arising when retailers crowdsource last-mile deliveries to their own customers, who can refuse or accept in exchange for reward. A planner must identify which offer, knowing that all need fulfilment, either via crowdsourcing using retailer’s vehicle. formalise problem and position it both literature about among routing problems not customers visit. show evaluate objective function this stochastic even one solution, needs solve an exponential number Problems. To address complexity, we propose Machine Learning Monte Carlo simulation methods approximate function, branch-and-bound algorithm heuristics reduce evaluations. these approaches work well on small size instances derive managerial insights economic environmental benefits customers.