How can Lean Six Sigma help Machine Learning?

Note that this article was submitted and accepted by KDnuggest, the most popular blog site about machine learning and knowledge discovery.

I have been using Lean Six Sigma (LSS) to improve business processes for the past 10+ year and am very satisfied with its benefits. Recently, I’ve been working with a consulting firm and a software vendor to implement a machine learning (ML) model to predict remaining useful life (RUL) of service parts. The result which I feel most frustrated is the low accuracy of the resulting model. As shown below, if people measure the deviation as the absolute difference between the actual part life and the predicted one, the resulting model has 127, 60, and 36 days of average deviation for the selected 3 parts. I could not understand why the deviations are so large with machine learning.

After working with the consultants and data scientists, it appears that they can improve the deviation only by 10%. This puzzles me a lot. I thought machine learning is a great new tool to make forecast simple and quick, but I did not expect it could have such large deviation. To me, such deviation, even after the 10% improvement, still renders the forecast useless to the business owners.

Read more at How can Lean Six Sigma help Machine Learning?

Leave your comments below and subscribe us to get updates in your inbox.

Share on FacebookShare on Google+Share on LinkedInTweet about this on TwitterEmail this to someone

2015 New Year’s Resolutions for the Supply Chain Industry

2015 New Year’s Resolutions for the Supply Chain Industry

Resolution #1 – Stop using the term VISIBILITY

People say that information is power. I beg to differ. I say, an informed decision is power. The visibility term has been over used. I’ve even heard some say that getting visibility to your supply chain is 80% of the challenge. They must not have run a supply chain. I see many supply chain leaders that have visibility, some in excel and some in automated tools. The ones that don’t have visibility can easily call the supplier and get it. Getting visibility isn’t the challenge. The real 80% challenge is “what are you doing with the visibility?”

Resolution #2 – Read only ONE “Cool Theme” report

In 2015, I resolve to read only one Cool Theme report. I’m tired of research analysts peddling these themes as a means to gain an edge on readership. Yet, I watch the audience during some of these Cool Theme presentations. And, half the people are on their smartphone working core issues back home, while the Analyst is talking about how supply chains should save the Panamanian golden frog, reduce the ozone layer, produce products with plastic wire from 3D printers and generate forecasts from Facebook posts!

Resolution #3 – Stop moaning about Bad Data

Let’s face it, everyone has some form of bad data. And, when you include all your tiered suppliers, they have bad data. The one constant is that you will never fix all the internal and external bad data. Yet, I still hear supply chain leaders say they need to focus first on fixing the data. I’ve seen many presentations from “Top 25” supply chains and how they’ve cleaned data, and why they should be considered a top tier supply chain story.

Resolution #4 – Fix the Disruption you can influence, not the Disruption you are concerned with

There are two types of disruptions. That which you are concerned with, and that which you can influence.

Volatility, regulation, geopolitics, economics, energy, and the list goes on. These are in your Circle of Concern. They happen, and you should be concerned. Yet, many supply chain leaders face fail to focus on the Circle of Influence, the area where you can make a difference.

Resolution #5 – Scrap the Talent Research, Make Planners more Productive

After reading all the Talent Research done in 2014, the topics of attrition, retiring professionals, and university-business alignment, I notice a big gap. The one thing missing in all this Supply Chain Talent research is the concept of being more productive with the talent you already have.

How can every supply chain improve productivity? In every supply chain I’ve seen in my past 25 years, there’s one constant – they all use some form of Excel – mostly to search for exceptions. Planners spend half their day dumping ERP and BI data into Excel, and then search for exceptions.

What are your resolutions? Share with us by leaving comments or contact us for a discussion.

Share on FacebookShare on Google+Share on LinkedInTweet about this on TwitterEmail this to someone