Exact matching of attractive shared rides (ExMAS) for system-wide strategic evaluations

• DOI: 10.1016/j.trb.2020.06.006
• Publication Date: 2020-07-03T02:49:52Z
• Authors: Rafał Kucharski; Oded Cats
• Keywords: Leverage (statistics); Benchmarking

The premise of ride-sharing is that service providers can offer a discount, so travellers are compensated for prolonged travel times and induced discomfort, while still increasing their revenues. While recently proposed real-time solutions support online operations, algorithms to perform strategic system-wide evaluations crucially needed. We propose an exact, replicable demand-, rather than supply-driven algorithm matching trips into shared rides. leverage on delimiting our search attractive rides only, which, coupled with directed shareability multi-graph representation efficient graph searches predetermined node sequence, narrows the (otherwise exploding) search-space effectively enough derive exact solution. utility-based formulation paves way model integration in demand models, allowing cross-scenario sensitivity analysis, including pricing strategies regulation policies. apply series experiments case Amsterdam, where we analysis performance terms both computations under alternative demand, network settings as well behavioural parameters. In 3000 offered 30% discount form 1900 achieving average occupancy 1.67 yielding vehicle-hours reduction at cost halving provider revenues 17% increase passenger-hours. Benchmarking against time-window constrained approaches reveals reduces more effectively, substantially travellers.