Originally posted by: EGU2_Jim_Banigan

We are a technology company that assists Distributors of drugs to analyze and award offers from pharmaceutical manufacturers for generic drugs. As you might imagine there are typically 20 or so manufacturers for a given generic drug. Each of which will submit offers that include pricing discount tiers. Tiers are based upon attainment of purchase thresholds which are measured quarterly. The larger the manufacturer, the more likely the discount tiers will be based upon aggregate purchases of total manfacturer sales across their entire product catalog. Offers tend to be bundled and tiered, thus binding the purchases of one drug to another; the sum of which is measured to determine achievement of a particular price threshold or tier. The dynamic nature of such pricing makes awarding the lowest priced offer for a particular targeted product complex. It is possible therefore that awarding a drug may enable the distributor to qualify for a lower pricing tier causing their purchase price for a host of products to go down. Conversely, shifting a particular product award from one manufacturer to another may cause the pricing tier for the losing manufacturer to go up, causing a price hike across a host of other manufacturer products. These implications need to be carefully measured when making a determination of what an optimal blend of awards is.

Our contracts platform houses all manufacturer offers and awards as well as historic purchases for a given distributor. The algorithm that we are seeking to integrate into our contracting platform should permit the distributor to identify a set of targeted drugs, The algorithm would then aggregate all known offers and awards for those targeted drugs and explore all possible combinations of awards to determine what would be the most optimal blend.

Is anyone familiar with a solution that has been created that could readily accomodate optimization of awards? Can anyone recommend a person that might be able to develop this algorithm? Need a rainmaker here
#DecisionOptimization
#MathematicalProgramming-General

Originally posted by: SystemAdmin

I can't point to any specific literature (and I have no idea if there are commercial software products for this problem), but the combination of quantity discounts and product bundles certainly gets discussed in the procurement/operations management literature, and I may have assigned something like this as a homework problem once upon a time. If you assume that volumes to be purchased are known with certainty (and that's a big "if"), and if you assume that time horizons for discounts are the same (e.g., all vendors base their discounts on calendar year aggregate purchases, or month-to-month aggregate purchases), I don't think the model is too complicated.

The specifics depend on things like all-units v. incremental discounts, whether the buyer insists on single-sourcing every item (or, alternatively, insists on at least two sources for each item), etc. How hard the model will be to solve is an empirical matter (i.e., try it and see), and obviously depends on how many vendors, how many products and how many bundles are involved (along with planetary alignments, solar flare activity, ...).

Paul

Mathematicians are like Frenchmen: whenever you say something to them, they translate it into their own language, and at once it is something entirely different. (Goethe)
#DecisionOptimization
#MathematicalProgramming-General

Originally posted by: EGU2_Jim_Banigan

Thank you Paul for your response. If I was seeking a consultant to provide input on the development of this application would you be available. If so, please let me know what your rate would be.
#DecisionOptimization
#MathematicalProgramming-General

Originally posted by: SystemAdmin

Might be best to move this off the forum. You can email me directly at rubin <at> msu <dot> edu.

Paul

Mathematicians are like Frenchmen: whenever you say something to them, they translate it into their own language, and at once it is something entirely different. (Goethe)
#DecisionOptimization
#MathematicalProgramming-General