Optimal dynamic offer for stochastic occasional drivers

The last-mile costs of deliveries can represent 50\% of total costs in online sales. To reduce this cost, some companies use crowdsourcing, inviting occasional couriers (EOs) to contribute to deliveries, reducing the use of their professional fleet (FP). We propose a stochastic-dynamic model assuming that EOs arrive at the store sequentially and their probability of accepting a delivery depends on the offered value, modeled by a sigmoid function. The objective is to determine the optimal offer values for each EO in order to minimize the total expected cost of deliveries. The expressions presented can be used to generate a table of optimal offers. In this extended abstract, we present only the cases where the cost of using the FP is proportional to the number of deliveries.