HighlightsNew VRP constraints including last-in, first-out deliveries and variable capacity.Metaheuristic that combines GRASP, tabu search and large neighborhood search.Solutions to real instances with over 150 stores, 500 orders, and 130 vehicles.Extensive benchmarking against operational routes and column generation algorithm.Cost reductions of $6429 or 6.72% per day, on average.
AbstractThe retail route design problem extends the capacitated vehicle routing problem with time windows by introducing several operational constraints, including order loading and delivery restrictions (last-in, first-out), order-dependent vehicle capacity, material handling limits at the warehouse, backhauling, and driving time bounds. In this paper, the problem is modeled on a directed network for an application associated with a major grocery chain. Because the corresponding mixed-integer program proved too difficult to solve with commercial software for real instances, we developed a greedy randomized adaptive search procedure (GRASP) augmented with tabu search to provide solutions. Testing was done using data sets provided Kroger, the largest grocery chain in the US, and benchmarked against a previously developed column generation algorithm. The results showed that cost reductions of $4887 per day or 5.58% per day on average, compared to Kroger’s corresponding solutions.
Route design for last-in, first-out deliveries with backhauling
Tramsportation Research, Part C: Emerging Technologies ; 76 ; 90-117
2017-01-05
28 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Route design for last-in, first-out deliveries with backhauling
Online Contents | 2017
|Truck backhauling on networks with many terminals
Elsevier | 1986
|Efficient GSM/UMTS Backhauling over Satellite
AIAA | 2006
|Truck backhauling on two terminal networks
Elsevier | 1984
|Recent Innovation in Last Mile Deliveries
Springer Verlag | 2014
|