This
thesis considers a two-item two-warehouse periodic review inventory model
that allows transshipment between warehouses. Transshipment decision between
warehouses is dependent on the time to the next order and the state of
inventory in the other warehouse when a stock out occurs in a warehouse. The
objective function considered is the total operating cost which comprises the
variable ordering costs, the holding costs at the warehouses, the costs of
transshipment between warehouses and the cost of emergency orders if
transshipment is not possible. An innite horizon dynamic programming model is
used to develop the objective function. As the resulting optimization problem
is a non-linear integer programming problem, we propose a heuristic to solve
the problem. The proposed heuristic is a combination of Greedy heuristics and
Lagrangian relaxation methods. The advantage of the Lagrangian method is its
ability to provide a test for the quality of the solution. A series of
numerical experiments performed not only illustrates the method proposed but
also shows that optimality can be achieved using the proposed heuristic.
Further, the method also provides the optimal instants when emergency orders
will be preferred over transshipment.
|
