Last mile logistics in mega-cities for perishable fruits

Purpose: A common problem in mega cities is congestion, due to the size of the automotive park, this makes that the perishable foods decreasing their organoleptic characteristics or increases their losses, which requires considering the effect of time on routing problems. The state of the art demonstrates the need to formulate new routing models that include the specific characteristics of perishable foods in order to reduce their losses. Design/methodology/approach: A mathematical model was formulated based on two classical models: the three-index vehicle flow model proposed by (Golden, Assad, Levy & Gheysens, 1984) and the time window model proposed by (Cordeau, Desaulniers, Desrosiers, Solomon & Soumis, 1999). We proposed a novel VRP Model that permits reductions loss due to the perishable. Findings: The optimum cost is found with AMP® for twenty nodes, six vehicles and six fruits. For more nodes, a two-phase strategy is proposed, first a clustering based on a modified p-median model and then a VRP for each cluster. Research limitations/implications: The results showed the need to investigate multi-objective models, since the performance measures can be efficiency, quality and response capacity; the model can be applied in other supply chains of perishable foods. Social implications: According to FAO in Logistics practices in the last mile generate between 10-30% of the perishable food loss in developing countries' mega-cities. Originality/value: A last-mile logistics strategy is proposed to manage delivery routes for fresh fruits in mega-cities, considering the effect of congestion through travel time in the perishability function. The new model it uses the flow variable to control the amount of each fruit arriving to each node and the time variable to define fruit waste or loss depending on the time and type of fruit.