Build-pack planning for hard disk drive assembly with approved vendor matrices and stochastic demands

We study a production planning problem based on hard-disk drive assembly, where the customers choose their preferred suppliers for pairs of inter-dependent components through the approved vendor matrix. The problem is to develop production plans that minimize the expected total shortage and holding costs while observing the matrix restrictions and limited components supplies. We provide a mixed-integer formulation of the problem, whose linear approximation and relaxation is solved using the column generation method. This is then embedded in a branch-and-price framework to solve the mixed-integer problem. Special branching rules are devised to preserve the structure of the pricing sub-problem at each branch-and-bound node. Finally we present some computational results of our implementation.