I am trying to find some special structure of a type of multidimensional knapsack problem, in which all the capacities of dimensions are the same. so the capacity constraint can be formulated in a piecewise linear form. My question is that will this structure help solve MKP? Because I think it is well known it is much difficult to solve. asked 19 Jun '13, 22:18 jc W 
What do you mean with "all the capacities of dimensions are the same"? Do you mean that entries of the resource vector b (in your knapsack constraints Ax <= b) are all the same real number? In my experience MKPs are difficult to solve only if you insist on obtaining "exact" solutions. For most practical purposes you can set solver tolerances to 15% acceptable optimality gap, and commercial solvers will return such solutions basically instantly. EDIT: the following is a fairly recent survey on methods to solve the knapsack and its variations, you may use it to get some initial reference if you intend to go deeper in this direction despite the name of the file it is concerned with the quadratic version only marginally actually answered 26 Jun '13, 06:25 Robin V yes u r right i men every entry in vector b is the same. Thx for the survey. I read that before. My question here is definitely on exact optimality. Of course in reality can try some metaheuristic to get good solutions (But i think that still depends on the scale like several hundred variables and less than 50 dimensions). But as I said, I am asking for some insight that how to utilize the feature or is there some structure I can use. For instance, will that help in doing reductions?
(26 Jun '13, 07:31)
jc W
