Daily Archives: April 12, 2005

Six Sigma I: An Introduction

Originally posted on on the e-Sourcing Forum [WayBackMachine] on Friday, 1 September 2006

The design of responsive supply chains is becoming a priority for companies who must continually reduce costs and streamline processes to remain competitive in a dynamic global market where companies that do not anticipate challenges and prepare for the unexpected can vanish almost overnight. This means that an organization today must pay more attention to the business environment, get tighter with its customers, analyze performance data better, and avoid disasters.

Six Sigma, which defines itself as a relentless quest for perfection through the disciplined use of fact-based, data-driven, decision-making methodology, is one methodology that companies can use to make their supply chains more responsive, foster innovation, and improve quality across the board. When properly applied, this ultimately leads to lower costs, greater profits, greater customer satisfaction rates and earnings per share for the shareholders, which is why Six Sigma has caught on at a number of large enterprises.

But it’s no light undertaking – it requires massive commitment from the CEO down, considerable amounts of training and application, master black belts, black belts, and green belts, and operational changes across the board. In addition, it often requires new mindsets, new methodologies, new technologies, new performance metrics, and, most importantly, new incentives. Which leads one to ask, if you’re not a large enterprise with a large spend, large budgets, and the time to transform, is it worth it?

That’s not an easy question to answer. If you just look at the average results, you might think it is a resounding yes. But no decision is ever that easy, especially when you do not really understand what something is and separate the facts from the hype.

After all, we are dealing with a methodology that is promoted as a revolution capable of fixing everything wrong with your corporation, with the possible exception of the clog in the lunch-room sink. (No wait, it has a process to fix that too!) Just one article I read recently said Six Sigma will help you with:

  • Globalization
  • Mergers and Acquisitions (M&A)
  • Supply Chain Design and Planning
  • Customer Relationship Management (CRM)
  • Brand Building
  • New Product Development (NPD)
  • Sustainable Growth
  • Innovation Management
  • Risk Management
  • and much, much more by
    • accelerating a company’s learning cycle
    • speeding up research phases
    • speeding up redesign
    • enabling rapid information exchange
    • … and so on.

To fully understand the breadth of this claim, imagine if a smart looking guy in a suit walked into your office and said “I’m the architect, foreman, electrician, plumber, steel worker, heavy equipment operator, drywaller, and project manager for that new office building you’re thinking of. With just a few humble assistants, I’m going to take on that massive project all by myself and finish it faster, better, and at less cost then any of the multi-billion contracting firms you are currently considering.” That’s comparable to the breadth of the claim that Six Sigma often appears to be making.

Theoretically, a universally applicable best practice generic methodology could be applicable to each of your business functions, but just how helpful is it going to be if it is that generic? That’s one of the questions we are going to try to answer this weekend. To do that, we are first going to focus on its potential contributions to quality and innovation, since these are two common qualities that define great operations across organizational functions. Sunday we will discuss its potential value to strategic sourcing.

However, first of all we are going to define what Six Sigma really is. Sigma measures variation, and more specifically, the standard deviation of a data set. Six Sigma refers to a distance of six standard deviations between the mean value of the data set and the nearest specified tolerance limit. A Six Sigma Process is one that produces at most 3 defects per million trials (in the long one).

Furthermore, Six Sigma is concerned with something called the Rolled Throughput Yield of a system, which ensures that the measurement applies to the finished product and not just a step in the process. The rolled throughput yield is calculated by multiplying the yield of each step. For example, a system with five steps and only 99% yield at each step would produce a rolled throughput yield of 0.99^5 = 0.951 or 95.1%. Thus, a six sigma process is one with a rolled throughput yield of at least 99.9997% or almost six nines reliability.

In short, a Six Sigma methodology is one that is designed to eliminate variation from a process to ensure consistent results every time.

For more information on six-sigma, see the “Six Sigma: Improve Supply Chains through Methodology” wiki-paper over on the e-Sourcing Wiki [WayBackMachine].