Six Sigma Intro 1

Six Sigma is a problem solving methodology for improving business / operational performance; it’s about minimizing mistakes and maximizing value. Traditionally Six Sigma was seen as a quality improvement methodology but today, it’s evolved to be much more than that.

From a statistical standpoint, Six Sigma aims to achieve 3.4 defects per million opportunities for defects. That means, if your process runs and outputs widgets; 999,996. 60 of those million widgets need to be defect free.

Why would we look to reduce defects to that level? Well, research suggests that many companies spend 25-40% of revenue fixing mistakes / re-doing processes. This is why many organizations choose to implement Six Sigma.

To implement Six Sigma we need to look at two levels. The managerial level and the technical level.

From a managerial standpoint, we use Six Sigma to achieve predictability and control in a business / a process. We do this by applying science through Six Sigma to the process in order to ascertain how parts, materials, machines and people should work together to output widgets at the highest speed and lowest cost.

By achieving faster and lower cost output, we have a direct, measurable financial impact, which leads to an easy justification of the Return on Investment (ROI) of Six Sigma initiatives.

A successful Six Sigma strategy / organizational infrastructure is comprised of:

A documented plan, which will include the strategic focus & business goals of the Six Sigma initiative. It’ll include the deployment plans; implementation schedules; activity tracking strategy & reporting techniques to be utilized.

Next, we have the strategy method – how we’re going to execute the strategy. That includes training the organization on Six Sigma; introducing Six Sigma champions, Black Belt, Green Belt and Yellow Belt Six Sigma professionals into the company to drive the initiatives forward.

Component three includes various plans and models. Which includes a competency model; role descriptions (for the SS project) and compensation plans.

We then have project definition guidelines which includes project savings criteria and guidelines; forecasting and validating project savings and the audit, evaluation and reporting of project savings guidelines.

After this, we need to define criteria for selecting Six Sigma projects. This includes defining project types; developing problem definition statements; approving and managing projects.

The IT strategy is the next cornerstone of our Six Sigma strategy. This will define the procedures and dashboards we require; the tools for designing & managing process models; tools for tracking projects; reporting on data and carrying out analytics.

Then we have the communications strategy, which defines how we’re going to disseminate the Six Sigma strategy & communicate educational content to the business.

Finally, we have the management review process which looks from top to bottom at Six Sigma effectiveness. The top is an aggregate of all Six Sigma projects running across the business. The middle level is an aggregated view of all Six Sigma projects within a business unity and the bottom view shows the individual projects.

So, now we’ve justified Six Sigma and understood the items that comprise a successful Six Sigma architecture, we need to ensure that we’re delivering what we should be delivering – i.e. what the customer actually wants. No process or business improvements can occur without understanding the customer, which we do through numerous VOC (Voice Of the Customer) tools.

A very simple tool we can use is a specification. To understand specifications we can think about Cadburys. When you buy a chocolate bar from Cadbury, you expect it to taste identical to the last one you had & you expect the texture to be the same. To achieve this consistency, Cadbury have a specification which defines what must be delivered for the customer to be satisfied.

We have various specification types:

  • One sided specifications – which simply provide a MAX or a MIN value
  • Two sided specifications – which provide a MAX and a MIN
  • An upper specification – which provides only a MAX
  • A lower specification – which provides only a MIN
  • Target – a single value to which the process should perform

This helps us generate acceptable tolerances, like the below:

So, if we look at this from the perspective of a coffee shop. If you buy a coffee, they want to deliver it somewhere in a range of temperatures. Before we set this, we need to use the RUMBA framework to keep our limits realistic.

They need to be kept realistic because, if the limits are too tight, you’ll expend huge amounts of capital resource making them happen. If they’re too lax, you’ll have upset customers. So, our coffee company setting limits of between 100 Celsius and 100.5 Celsius for their coffee by the time it reaches the customers table may not be realistic. Whereas, a range of 95 Celsius to 105 Celsius will be easier to achieve & still ensure customer satisfaction.

So, Rumba stands for:

  • Reasonable – realistic assessment of customers’ needs
  • Understandable – clear and unambiguous requirements
  • Measurable – can we measure that our target is it?
  • Believable – can it happen in the environment we’re in?
  • Achievable – Is it possible?

So, now that we’ve got this, we need to define what quality actually is. So let’s look at the below diagram. We’ve produced 4 new screws which must be between 3MM and 3.1MM in diameter. The yellow is significantly below the lower limit, so it’s certainly our lowest quality output. The black is exactly on target, so it’s our highest quality unit.

We can define quality as the ‘on target performance, with as little variation as possible’. So the black dot below has absolutely no variance from our target & is therefore definitely our highest quality unit.

We need to ensure quality as our costs increase with any deviation from the target. So we could view it like this (where cost is the Y axis).

To achieve maximum productivity and lowest costs, we need to ensure that all of our products cluster around the target, as those towards either limit will incur additional costs. Those costs aren’t always obvious either, but we have items such as:

  • Scrap
  • Design changes
  • Re-Work
  • Long cycle times
  • Excess inventory
  • Warranty costs
  • Lost loyalty from customers
  • Excess material orders
  • Lost sales
  • Rejected products

So, when we have lots of components, being produced independently that must work together, we need to establish extremely high probability of success for each of the individual components. Think of it like having dice. If you have one dice, there is a 1 in 6 chance of rolling a five. If you have 2 dice, there is a 1 in 12 chance of rolling a five. If you had 100 dice, you’d have a 1 in 83 million chance of NOT rolling a five.

So, in order to make a robust final product, we require Six Sigma on all of the individual components. If achieved, we can ensure exceptional performance when all products are assembled together.

Content based on study of the Six Sigma Black Belt course and Six Sigma for Dummies