This paper presents a solution approach for an inventory routing problem (IRP). An inbound material-collection
system is considered. It consists of a set of suppliers who produce non-identical items, and a central warehouse stocking a
number of unique items that face constant and deterministic demands from outside retailers. With an economic order
quantity (EOQ) inventory policy, the items are jointly replenished and collected by a fleet of identical vehicles that have
capacity and frequency constraints and are dispatched from the central warehouse. A greedy randomized adaptive search
procedure (GRASP) is proposed to solve the problem. GRASP is an iteration process that consists of two phases, a
construction phase and a local search phase. In the construction phase, an initial feasible solution is generated using the
greedy heuristic based on the Distance Sum heuristic. In the local search phase, a very large-scale neighborhood search
(VLSN) algorithm is applied to improve the solution. To measure the performance of GRASP with VLSN, computational
experiments are conducted on randomly generated problems and solutions are compared with the lower bound on the
total costs. The results reveal that the GRASP with VLSN performs efficiently in finding near-optimal solutions.