The following folders contain the instances and the results of our study on: A unified decomposition matheuristic for assembly, production and inventory routing By: Masoud Chitsaz, Jean-Francois Cordeau and Raf Jans GERAD, CIRRELT and HEC Montreal, Canada H3T 2A7 1- "Instance" folder: "ARP_Chitsaz" subfolder: ARP instances developed in our study including 1440 instances, "IRP_Archetti_2007" subfolder: Small IRP instances presented in Archetti et al. (2007) including 4 folders and a total of 160 single-vehicle instances, "IRP_Archetti_2012" subfolder: Large IRP instances presented in Archetti et al. (2012) including 60 single-vehicle instances, "PRP_Bertazzi" subfolder: PRP instances presented in Archetti et al. (2011) including 1440 single- and multi-vehicle instances, and "PRP_Boudia" subfolder: PRP instances presented in Boudia et al. (2005) and Boudia et al. (2007) including 90 multi-vehicle instances. 2- "Result-Paper" folder: The detail results are provided in the same folder names as in the "Instance" folder. In every ARP, IRP and PRP .txt solution file, the results are provided with the same format: The first two rows include some general information. The third row introduces some of the following information depending on the problem and instance specification: set: Name of folder/Number of nodes, ins_id or count: Instance ID, trans_cost: transportation cost update mechanism, 2: marginal cost updating, n: Number of nodes, l: Number of periods, Horizon: For IRP instances only, 1 or 2 to address the instances with 3 or 6 planning periods, respectively, inv_cost: For IRP instances only, 1 or 2 to address the instances with low or high inventory costs, respectively, m: Number of vehicles, Q: Vehicle capacity, C: production capacity, u: Unit production cost, f: Fixed setup cost, tot_time: Total time CCJ-DH spent to solve the instance, and sln_val: Best solution value obtained by CCJ-DH. The next row addresses the following information for the rest of the file: p_t: For the ARP and PRP, if production takes place, a non-zero value will apear at the beginning of each row. Otherwise, a 0 shows a period with no setup, r_0t: For the IRP, it shows the amount of initial inventory at the depot, x_ijt and q_it: The pairs of node number and quantity, respectively. It starts from depot by "0 0" and when ever the vehicle is returned to the depot a "0 0" will appear, and_so_on: a "-1" at the end of the row indicates that there is no more vehicle routes for this period. Note that: - every row contains the production (in case of PRP) and route(s) for one period, - having the production amount and shipment quantities, the inventory variables can be obtained using the flow constraints, and - for the ARP and PRP_Bertazzi, the ".01" right after the production amount should be ignored. References: Archetti, C., Bertazzi, L., Hertz, A. and Speranza, M. G. (2012). A hybrid heuristic for an inventory routing problem, INFORMS Journal on Computing 24(1): 101–116. Archetti, C., Bertazzi, L., Laporte, G. and Speranza, M. G. (2007). A branch-and-cut algorithm for a vendor-managed inventory-routing problem, Transportation Science 41(3): 382–391. Archetti, C., Bertazzi, L., Paletta, G. and Speranza, M. G. (2011). Analysis of the maximum level policy in a production-distribution system, Computers & Operations Research 38(12): 1731–1746. Boudia, M., Louly, M. A. O. and Prins, C. (2005). Combined optimization of production and distribution, CD-ROM Proceedings of the international conference on industrial engineering and systems management, IESM, Vol. 5. Boudia, M., Louly, M. A. O. and Prins, C. (2007). A reactive grasp and path relinking for a combined production– distribution problem, Computers & Operations Research 34(11): 3402–3419.