A joint optimisation model for designing demand responsive connectors fed by shared bikes

Demand Responsive Connectors (DRCs) have become a more general-purpose flexible transit service that caters to patrons' personal needs. The traditional DRC operation, however, suffers from low efficiency due to excessive detours and incurs diseconomies of scale with respect to the demand and service area. To tackle this issue, we propose a novel DRC fed by shared bikes, which functions as an access/egress mode for certain request points. Analytical models are derived for the joint design of such a hybrid system. A mixed-integer non-linear programme is established to minimise the total system cost. A heuristic solution algorithm is developed by combining the simulated annealing and branch-and-bound algorithms. A series of numerical cases are designed to evaluate the proposed system's performance against the traditional ones. The results demonstrate that the introduction of shared bikes can reduce the DRC tour length and consequently save total system costs.