Procurement Beyond ERPs : A Choice

In the environment of high volume procurement (such as commodity procurement), one particular factor clearly stands out. Multiple teams across different geographies often struggle with coherence (necessity!) and congruence (luxury!) in their procurement timing and across their commodities. It is a bit impractical to expect ERPs to facilitate such practices.
Majority of the procurement teams follow their own commodity cycles and price trends to plan the spending. This approach works well from micro management perspective.
However, in order to substantiate an overall procurement strategy (say, for upcoming quarter), it is extremely important to create a framework on macro basis.
You must have read about Permutation and Combination! I am taking the same concept and extending it into the world of Procurement. I have been experimenting with concept from last 6 months now. Results have been mixed so far. But, I am able to derive certain advantages which are just not possible using present day traditional practices.
Let me explain you the picture below!
Example, if we have to buy 4 commodities — A,B,C,D.
Either we can buy them in any order (Combination) or we can buy them in a particular sequence (Permutation). Most of the procurement managers (and teams across geographies) follow Combination driven procurement methodology.
Here is what I did to test Permutation driven procurement:
1.) I studied the pricing formulae & underlying cost drivers for each commodity
2.) From the cost driving parameters, I mapped out correlation matrices among common parameters.
3.) Using these correlation matrices, the started the Procurement simulation experiment. Following were my test objectives (For the sake of clarity, I am taking only two factors — Exchange Rate and Crude Price)
3a.) Ex rate increase and Crude price increase
3b.) Ex rate increase and Crude price decrease
3c.) Ex rate decrease and Crude price increase
3d.) Ex rate decrease and Crude price decrease
Then for each case, I took daily prices of each commodity index. I used the correlation matrices weights to evaluate which particular sequence of procurement (Commodity A,B,C,D) will given me lowest value of procurement expenditure.
To make the model steady, I opted for neural network instead of Random Forest algorithm.
The final output gave me a recommendation of commodity permutation for 3a,3b,3c and 3d scenario. I am excited to continue this experiment as I see huge potential for cost savings.
Conceptually, Permutation based procurement makes sense because if underlying cost drivers are common, it is an obvious deduction to time the procurement accordingly.
So, which type of procurement team do you manage? Permutation driven or Combination driven?
Do let me know your thoughts!
My motivation :
I love building stuff. I am combining my little procurement and supply chain knowledge with my machine learning skills to go beyond today’s procurement practices. I do not believe that going with best out of the 3 procurement style, need any skill. It can be and should be automated. Also, I have also seen the majority of manufacturing companies using ERP systems as transaction registers.
I am trying to create Procurement managers more Data Savvy where they focus on real value and not just transactions.
Contact me on gsharma.imbox@gmail.com for more details. I will be happy to assist.
