Search By Topic The Green Supply Chain Distribution Digest
Supply Chain Digest Logo

Dr. Michael Watson
Northwestern University

Supply Chain by Design

Dr. Michael Watson, one of the industry’s foremost experts on supply chain network design and advanced analytics, is a columnist and subject matter expert (SME) for Supply Chain Digest.

Dr. Watson, of Northwestern University, was the lead author of the just released book Supply Chain Network Design, co-authored with Sara Lewis, Peter Cacioppi, and Jay Jayaraman, all of IBM. (See Supply Chain Network Design – the Book.)

In addition to teaching at Northwestern, Watson is a founding partner at Opex Analytics. 

February 21, 2017

The Most Overlooked and Underestimated Data in Supply Chain Design

Building a Process to Maintain Dense Matrix of Data Will Provide Flexibility Within the Supply Chain Model


Anyone who has done or seriously thought about a network design study realizes that you need a lot of data.  You need data on your customers and their demand, on your different products, on where product comes from, and the costs to move product around in the network.  And, this is just part of the data.

People (myself included) make the mistake of overlooking the other part of the data-- the creation of the dense data matrix. 


Watson Says...

...This data is the toughest to get and you don’t realize that you don’t have it until after the baseline model is built. 

What do you say?

Click here to send us your comments

What is a dense matrix of data?  We are using this term to refer to filling in the data for all possibilities-- what you do today and what you want to consider.


For example, the dense matrix of data for transportation is the cost to make any possible move in the supply chain.  In production, it is the cost to produce any item at any plant.  In warehousing, it is the cost to handle and store any item, no matter where product comes from our goes to.


This sounds obvious, but surprisingly it isn’t.  And, it isn’t obvious because this data is the toughest to get and you don’t realize that you don’t have it until after the baseline model is built.


This data is toughest to get because you don’t already have it and you have to match it to your existing data. 

For example, you likely have transportation rates from your existing warehouses to the customers you currently service.  The rates have been thoroughly negotiated and likely reflect market conditions.  When you ask a carrier for a rate for a lane that you aren’t currently using, you are likely to get a non-negotiated rate.  You have a few choices, you can use all non-negotiated rates in your model or you can use statistical techniques to figure out how well you would negotiate various lanes.


With production, it can be even harder.  Your standard costs show you how much it costs to make an item in a given plant.  But, this is an accounting cost.  It likely has fixed costs baked into it and has some assumption on how much of that product you are likely to make.  The result of your supply chain design model may double or half the amount a plant makes and change the product mix while it is at it.  You need your dense matrix of costs to be prepared for this.


On top of the difficulty, this data is often easy to overlook.  Usually a first step in the model building practice is to get a baseline model up and running.  The baseline model is meant to replicate your existing supply chain.  You use this model as one step in the validation process.  To build this model, you don’t need the dense matrix of data.  You just need data for things are you already doing.


If you haven’t built the dense matrix of data, you are going to be quickly disappointed when you try to run some initial scenarios and realize you need to go collect this data.


Final Thoughts


If you want to maintain the ability answer what-if questions with your supply chain model, you need to build a process to maintain the dense matrix of data.  This will give you a lot of flexibility.

Any reaction to this Expert Insight column? Send below.

Your Comments/Feedback

Erik Lopezmalo

Sr. Operations Research Analyst, Land O' Lakes
Posted on: Feb, 21 2017
Great article. Sometimes we modelers tend to fill out data gaps with "educated guesses / assumptions" on how cost components would behave due to lack of more realistic information. Part of the process to mantain the dense matrix of data should include the updating / maintaining of those assumptions.

john fitzgerald

Dir, GT Nexus
Posted on: Feb, 24 2017
 I agree 100%. To truly look at end to end supply chain total landed costs you need all transportation costs including transport plus customs for all modes  including Ocean, dray, rail, truck from the first mile to last mile. This data needs to be cleansed and normlalized if it is to be used at all and gathering it as it happens near real time is already happening--Remember "little data eats big data for lunch."

Clifford Blackburn

Mr, TSI Central Station
Posted on: Mar, 03 2017
There are so many articles since the beginning of 2017 that speak exactly to what you are saying. The supply chain world is changing, the revolution has begun. Data is everything, having the ability to collect, manage and distribute your data is what will keep you ahead of the croud. Great article, thank you.



Follow Us

Supply Chain Digest news is available via RSS
RSS facebook twitter youtube
bloglines my yahoo
news gator


Subscribe to our insightful weekly newsletter. Get immediate access to premium contents. Its's easy and free
Enter your email below to subscribe:
Join the thousands of supply chain, logistics, technology and marketing professionals who rely on Supply Chain Digest for the best in insight, news, tools, opinion, education and solution.
Home | Subscribe | Advertise | Contact Us | Sitemap | Privacy Policy
© Supply Chain Digest 2006-2023 - All rights reserved