SEARCH searchBY TOPIC
right_division Green SCM Distribution
Bookmark us
sitemap
SCDigest Logo

SCDigest Expert Insight: Supply Chain by Design

About the Author

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.)

Prior to his current role at Northwestern, Watson was a key manager in IBM's network optimization group. In addition to his roles at IBM and now at Northwestern, Watson is director of The Optimization and Analytics Group.

By Dr. Michael Watson

November 13, 2012



Advanced Analytics in Supply Chain - What is it, and is it Better than Non-Advanced Analytics?

Better Defining the Field of Analytics by Breaking it Down into Three Categories


We've had readers ask what is "advanced analytics?" If you logically answer this question, you also have to define "non-advanced analytics."

Dr. Watson Says:

start
When you are evaluating analytics solutions, you should understand whether the solution is descriptive, predictive, or prescriptive.
close
What Do You Say?



Click Here to Send Us Your Comments
feedback
Click Here to See Reader Feedback


To no one's surprise, you don't see many vendors talking about their great "non-advanced" analytics solution or managers proposing a "non-advanced analytics" project to the CEO.

This discussion highlights that although the term analytics is widely used, it is very poorly defined. And, a poorly defined word with such good connotations is in danger of becoming a buzzword - vendors call everything they do "analytics" and managers put the word "analytics" in all their projects.

So, before we get into "advanced" analytics, we should define analytics. If we go back to the Davenport’s "Competing on Analytics" Harvard Business Review article that kicked off the analytics movement, he defines analytics as "the ability to collect, analyze, and act on data."

In other words, at a high level, analytics is the ability to use data to make better decisions.

Unfortunately, this does not help us much. Haven’t companies always tried to use data to make decisions? - Yes, they have. Aren't there thousands of ways to analyze data? Yes, there are.

No wonder people are confused.

Fortunately, academic and professional organizations have realized that the field of analytics should be broken down into three categories:

1. Descriptive analytics-- using historical data to describe the business. This is usually associated with Business Intelligence (BI) or visibility systems. In supply chain, you use descriptive analytics to better understand your historical demand patterns, to understand how product flows through your supply chain, and to understand when a shipment might be late.

2. Predictive analytics-- using data to predict trends and patterns. This is commonly associated with statistics. In the supply chain, you use predictive analytics to forecast future demand or to forecast the price of fuel.

3. Prescriptive analytics-- using data to suggest the optimal solution. This is commonly associated with optimization. In the supply chain, you use prescriptive analytics to set your inventory levels, schedule your plants, or route your trucks.




Columns by Dr. Watson

Supply Chain by Design: Four Lessons from Hau Lee's Green Car Story Updated for the Era of Machine Learning

Supply Chain by Design: Profitability of Your Assets Depends on how you use Them

Supply Chain by Design: The Most Overlooked and Underestimated Data in Supply Chain Design

Supply Chain by Design: Self Driving Trucks May Create New Roles and New Types of Jobs

Supply Chain by Design: Why Driverless Trucks May Create the Need for More Drivers

Supply Chain by Design: A Non Obvious Way Self Driving Trucks May Impact Your Network Design Strategy

Supply Chain by Design: How You Should be Using Multi-Echelon Inventory Tools

Supply Chain by Design: You Don't Need the Optimization in Multi-Echelon Inventory Optimization

Supply Chain by Design: On Network Modeling - Blaspheming the Baseline

Supply Chain by Design: Profit Maximization Feature and Amazon’s Focus on Lead Time to Grow Revenue

Supply Chain by Design: Using Profit Maximization to Minimize Cost

Supply Chain by Design: Two Big Reasons You Don't Want to Maximize Profit in your Supply Chain Model

Supply Chain by Design: You Can Set Inventory Levels and Other Such Myths

Supply Chain by Design: Six Organizational Issues in Tactical Inventory Planning

Supply Chain by Design: Four Things Nick Saban Can Teach us About Inventory Planning

Supply Chain by Design: Seven Ways You Can Think About Christmas Capacity to Avoid Having the Press Blame Your Supply Chain for Missed Deliveries

Supply Chain by Design: Machine Learning and High Quality Potato Chips

Supply Chain by Design: Simple Models to Solve Complex Fast Delivery Problems

Supply Chain by Design: CSCMP Report - Five Take-Aways on Natural Gas Trucks from the AB InBev and Owens Corning Talk

Supply Chain by Design: CSCMP Report - Six Insights from LLamasoft and JLL

Supply Chain by Design: CSCMP Report - Five Supply Chain Design Lessons from Benjamin Moore and How One is Used by Amazon in their 1-hour Delivery Service

Supply Chain by Design: CSCMP Report - Two Supply Chain Design Lessons from Starbucks CEO

Supply Chain by Design: Four Groups that Need to Step Up to Help Make Network Design More of a Profession

Supply Chain by Design: Top Four Best Practices for Multi-Objective Optimization

Supply Chain by Design: Caesars Entertainment's Customer Data is Worth $1B - How Much is Your Supply Chain Data Worth?

Supply Chain by Design: Using Transactional Data to Estimate Truckload Market Conditions in Near-Real-Time

Supply Chain by Design: Routing Optimization is Hard: Lessons from UPS

Supply Chain by Design: Eli Goldratt's Book "The Goal" is on Jeff Bezos (Amazon) Reading List

Supply Chain by Design: Top Five Rules for Cleaning Data for a Strategic Analysis

Supply Chain by Design: Network Design and Accounting Data

Supply Chain by Design: Sears Case Study - Same Day Delivery

Supply Chain by Design: Three Ways UPS and FedEx Handled Christmas 2014

Supply Chain by Design: Deepen Your Optimization Knowledge Over the Holidays with Free E-book

Supply Chain by Design: Calculating Supplier Lead-Time Variability - Not as Easy as It Seems

Supply Chain by Design: The Myth of the Market Rate and Network Design Projects

Supply Chain by Design: Three Ways the Supply Chain Wastes Big Data

Supply Chain by Design: Three Supply Chain Lessons from the book Scaling Up Excellence

Supply Chain by Design: What to Do About the Rise in Costs Because of the Trucker Shortage

Supply Chain by Design: Modeling Your Competitors

Supply Chain By Design: Just Because the Feature Exists, Doesn’t Mean You Should Use It

Supply Chain by Design: Just Because People are Talking about Big Data Doesn’t Mean it is Clean Data

Supply Chain By Design: Can Western Manufacturing Be Saved: What Does it Mean to Your Firm?

Supply Chain By Design: Optimized Baseline and the "Perfect" Network Design

Supply Chain By Design: Become more Analytics-Driven to Recruit Talent

Supply Chain By Design: Step Up Your Preventative Maintenance with Predictive Analytics

Supply Chain By Design: Demystifying Stochastic Optimization

Supply Chain By Design: More on Big Data in the Supply Chain

Supply Chain by Design: Should You Extend Your Network Design Capability with a Map Portal?

Supply Chain by Design: A New Trend in Network Design: Flow Path Modeling

Supply Chain By Design: Top 5 Skills You Need in a Supply Chain Modeler

Supply Chain By Design: Comment on Biggest Supply Chain Planning Technology Challenges

Supply Chain by Design: Beyond the Square Root of N Rule

Supply Chain by Design: UPS's Christmas Problem Explained in One Graph

Supply Chain By Design: Your One Network Design New Year's Resolution

Supply Chain By Design: 80/20 Rule for Supply Chain Design

Supply Chain By Design: What Makes a Good Inventory Buffer

Supply Chain by Design: Some Things Do Not Change: Cost and Service Trade-Offs with Air Shipments

Supply Chain By Design: The Impact of Natural Gas Trucks On Your Supply Chain Design and Capabilities

Supply Chain By Design: Controlling Inbound Transportation with Inventory

Supply Chain by Design: Three Types of Supply Chain Buffers

Supply Chain by Design: Systems Thinking and the "Limits" of Optimization

Supply Chain By Design: 3D Printing and Robotics - Disrupting the Dominant Supply Chain Model

Supply Chain by Design: Future Supply Chain- Airships and the Physical Internet

Supply Chain By Design: Avoiding Capital Investments - A Hidden Benefit of Network Design

Supply Chain by Design: Three More Reasons the Impact of the New Hours of Service Rules May Not Be So Drastic

Supply Chain By Design: Three Ways to Handle the Lag Time from Modeling to Implementation

Supply Chain By Design: Three Quick Steps for Analyzing Big Data

Supply Chain by Design: What is Big Data?

Supply Chain By Design: Using Optimization to Compare Facilities or Internal Benchmarking

Supply Chain By Design: Four Steps for Thinking About An Optimization Problem

Supply Chain By Design: Don't Let the Term "Optimization" Become a Buzzword

Supply Chain By Design: Supply Chain Models Can Go Wrong - A Different Perspective

Supply Chain By The Numbers: Top Three Ways Supply Chain Models Can Go Wrong

Supply Chain By Design: Supply- and Demand-Centered Modeling: A Follow Up to 2013 Priorities

Supply Chain By Design: Three SCDigest Predictions You Should Be Modeling

Supply Chain by Design: Cost to Serve Modeling

Supply Chain By Design: Top Five Models You Should Build in 2013

Supply Chain By Design: Should You Source from China or the US? Why Not Both?

Supply Chain By Design: Same Day Delivery and Network Design

Supply Chain By Design: Political Supply Chain and Network Design

Advanced Analytics in Supply Chain - What is it and is it better than Non-Advanced Analytics?

Having this definition gives you a better framework for evaluating analytics projects and understanding how they may help your supply chain. Note that this does not suggest that one type of analytics is better than another - different problems require different solutions.

Once we have this definition, we don’t need the generic term "Advanced Analytics." For various reasons, BI systems and some statistical solutions have become synonymous with the term analytics. So, to differentiate themselves, vendors offering optimization solutions, complex new statistical methods, or something that they thought was a breakthrough tried to label their solution as an "Advanced Analytics." Of course, once some vendors start using the term, others will follow.


Final Thoughts

When you are evaluating analytics solutions, you should understand whether the solution is descriptive, predictive, or prescriptive. Then, within each of these categories you can determine if the solution is rather basic or advanced and what will meet your needs.

Recent Feedback

Concise and on point without hyperbole! Thanks, Dr. Watson!


John Hill
Director
St. Onge Company
Nov, 16 2012

Nice final thoughts and I agree that analytics "is the ability to use data to make better decisions."

Another interesting aspect that we can draw from this (apart from what the data is telling us) is what the data is not telling us.

In fact, it is the "not-telling" that contributes to a greater degree of complexity.


Koh Niak Wu, Ph.D.
Global Supply Chain and Logistics
Dell Singapore
Nov, 27 2012

Great article Mike.  It may be too late to worry about "Analytics" becoming a buzzword.  Just like "analysis", "analyst" or "optimize" it's become part of the general vocabulary and has lost much of it's clarity.    I think I moved to calling what I do "Advanced Analytics" as a way of highlighting that analytics can't all be done in Excel.

For the people buying such software and services this is a crucial thing to understand - just because it says analytics on the box does not mean you will get anything more than simple reporting when you use it.  In fact, if my experience is at all representative, if its says "analytics" there is an excellent chance you will find nothing beyond reporting and perhaps visualization/alerting tools.   If what you need is a predictive model, you had better understand what constitutes "predictive analytics".

Personally I tend to use the "What happened?", "What if?" and What's Best"  categories to explain different sorts of analytics but perhaps it's time for me to make a change.


Andrew Gibson
Partner
Crabtree Analytics
Nov, 27 2012

This is really a good definition. The categories are very helpfull to support the managers in their Business Analytics projects. A clear definition about what optimizations they would like to describe, predict and suggest will demand a deep understanding about the capabilities they have in the transactional process to generate the required level of data and information that they will need to create an efficient and value added business analytics process.  So maybe they will realize that they need to start with some process improvements before starting a more sofisticated business analytics process.  


Valério Machado da Silva
Supply Chain Executive
Seeking for a new position.
Mar, 18 2013
 
.