School of Six Sigma

Six Sig­ma is a data-dri­ven approach to improv­ing qual­i­ty and effi­cien­cy in busi­ness process­es. Six Sig­ma aims for near-per­fect per­for­mance, tar­get­ing a goal of no more than 3.4 defects per mil­lion oppor­tu­ni­ties. The method­ol­o­gy employs the frame­work of DMA­IC (Define, Mea­sure, Ana­lyze, Improve, Con­trol) for deliv­er­ing improve­ment projects. Six Sig­ma prac­ti­tion­ers use sta­tis­ti­cal tools to iden­ti­fy root caus­es of defects and inef­fi­cien­cies, enabling pre­cise solu­tions that improve out­comes, increase cus­tomer sat­is­fac­tion, and con­tribute sig­nif­i­cant­ly to orga­ni­za­tion­al cost reduc­tion and profitability.

Now that we’ve cov­ered some impor­tant foun­da­tion­al work it’s time to learn how to iden­ti­fy a good project. This step is extreme­ly impor­tant since a poor­ly scoped project has lit­tle hope of success.

The abil­i­ty to under­stand how mate­r­i­al and process­es flow, while also iden­ti­fy­ing the ​“inputs” or ​“Xs” to a process, is at the heart of the DMA­IC method­ol­o­gy. So, in this sec­tion we’ll learn about dif­fer­ent types of process maps while also learn­ing how to take the infor­ma­tion learned from process map­ping into addi­tion­al tools like the C&E Matrix and FMEA in order to nar­row the lists of inputs down to the ​“crit­i­cal few.” To be sure, many prob­lems can be solved dur­ing this phase of a project, so it’s real­ly time to dig in!

Sta­tis­tics is the sci­ence of col­lect­ing, ana­lyz­ing, pre­sent­ing, and inter­pret­ing data. Sta­tis­tics pro­vides meth­ods for design­ing exper­i­ments, tech­niques for ana­lyz­ing data, and mak­ing informed deci­sions. Learn the dif­fer­ences between descrip­tive and infer­en­tial sta­tis­tics, and how to use each. This dis­ci­pline is essen­tial for enabling pro­fes­sion­als and researchers to eval­u­ate trends, test the­o­ries, and dri­ve improve­ment and inno­va­tion in any industry.

Accu­rate­ly inter­pret­ing data col­lect­ed dur­ing test­ing phas­es is cru­cial for enhanc­ing process­es and work­flows. Descrip­tive sta­tis­tics are sum­ma­ry mea­sures that describe, show, or sum­ma­rize the char­ac­ter­is­tics of data in a mean­ing­ful way. Ensur­ing that all team mem­bers are pro­fi­cient in descrip­tive sta­tis­tics and their prop­er appli­ca­tion, we can make informed deci­sions that dri­ve per­for­mance improvements. 

When it comes to see­ing how process­es change over time, the abil­i­ty to graph and make charts of data can be invalu­able, espe­cial­ly for con­tin­u­ous improve­ment prac­ti­tion­ers. Graphs and charts can even help empha­size dif­fer­ent attrib­ut­es of the same data. But how do we know if we’re using the best visu­als for our par­tic­u­lar data set? This course will help with that decision.

Hav­ing a sol­id under­stand­ing of the sci­en­tif­ic method and how to study a small sam­ple of data in order to make deci­sions about an entire pop­u­la­tion is what usu­al­ly sep­a­rates excel­lent Con­tin­u­ous Improve­ment Prac­ti­tion­ers from aver­age ones. In the next series of videos we real­ly dig into top­ics such as hypoth­e­sis test­ing and nor­mal­i­ty. Final­ly, you will nev­er need to use ALL of these tools in a sin­gle project… but it’s still impor­tant you under­stand the tool allow­ing you to use it when the time is right.

Sta­tis­ti­cal Process Con­trol, or SPC, is a method to mon­i­tor, con­trol, and enhance process­es. SPC enables us to build con­sis­tent, pre­dictable process­es by study­ing data over time and min­i­miz­ing vari­a­tion. Learn the five steps of Sta­tis­ti­cal Process Con­trol, how to cre­ate a sam­pling plan, set sub­group sizes and sam­ple sizes, the dif­fer­ence between attribute and vari­able data, how to select and cre­ate the appro­pri­ate type of Con­trol Chart for each sit­u­a­tion, and more.

Have you ever won­dered what ​“Six Sig­ma” real­ly means from a sta­tis­ti­cal per­spec­tive? If so, you’re about to learn. Addi­tion­al­ly, you’re also going to learn the key rela­tion­ship between the Voice of the Cus­tomer and the Voice of Process as well as how to mea­sure this rela­tion­ship with Process Capa­bil­i­ty Analysis.

ANO­VA, or Analy­sis of Vari­ance, is a method for com­par­ing the means of three or more sam­ples to deter­mine if at least one of the sam­ple means dif­fers sta­tis­ti­cal­ly from the oth­ers. Learn the key sta­tis­tics con­cepts of main effects, inter­ac­tion, influ­ence, degrees of free­dom, the F- sta­tis­tic, R‑squared, and how to fol­low the step-by-step process for con­duct­ing one-way and two-way ANOVAs.

When applic­a­ble, being able to deter­mine how ​“cor­re­lat­ed” vari­ables are can be extreme­ly enlight­en­ing as it can allow us to build pre­dic­tive mod­els. But, as you will learn, there is a key dif­fer­ence between cor­re­la­tion and cau­sa­tion so set­tle in and pre­pare to have some fun with regression!

The total vari­a­tion of a sys­tem is the sum of the actu­al vari­a­tion of a prod­uct, process, or sys­tem and the mea­sure­ment sys­tem vari­a­tion. MSA, or Mea­sure­ment Sys­tem Analy­sis, is a sta­tis­ti­cal method to ana­lyze whether the vari­a­tion with­in our mea­sure­ment sys­tem is accept­able. Learn the five char­ac­ter­is­tics of a mea­sure­ment sys­tem, the dif­fer­ence between attribute and vari­able data MSA, and the steps to con­duct and per­form a Gage R&R.

Design of Exper­i­ments, or DOE, is one of the most pow­er­ful tools avail­able to Lean & Six Sig­ma prac­ti­tion­ers. Sad­ly, many peo­ple sim­ply don’t under­stand what an authen­tic DOE is or, in some cas­es, some prac­ti­tion­ers mis­tak­en­ly believe their one fac­tor at a time exper­i­ment is in fact a DOE when, real­ly, it isn’t.

The final phase of a DMA­IC project is Con­trol. The pur­pose of this phase is to sus­tain the solu­tions and pre­vent prob­lem recur­rence. This requires tak­ing actions to update doc­u­ments and stan­dards, con­duct train­ing and com­mu­ni­ca­tion regard­ing the changes, and com­mit­ting to con­tin­u­ous mon­i­tor­ing and improve­ment of the process. Learn the key deliv­er­ables and activ­i­ties in the Con­trol Phase that enable this.

Fail­ure Modes and Effects Analy­sis, abbre­vi­at­ed FMEA, is a struc­tured approach for iden­ti­fy­ing the ways that a prod­uct or process can fail, what caus­es the fail­ure, and how this affects the cus­tomer. Learn how to apply FMEA to improv­ing the safe­ty, reli­a­bil­i­ty, and per­for­mance of process­es, prod­ucts, and services.

DMA­IC is a five-phase approach for design­ing, deliv­er­ing, and sus­tain­ing improve­ment projects. Learn the objec­tives of each phase, the struc­tured activ­i­ties of each phase, com­mon sta­tis­ti­cal tools and analy­sis tech­niques used dur­ing each phase, how to con­duct toll­gate reviews between each phase, and how project lead­ers must han­dle project closure.

The 8D prob­lem-solv­ing process was estab­lished in the ear­ly 1980s by the Ford Motor Com­pa­ny with the goal of stan­dard­iz­ing and improv­ing prob­lem-solv­ing capa­bil­i­ties in their sup­ply chain. Launched under the name of ​“Team Ori­ent­ed Prob­lem-Solv­ing Using the Eight Dis­ci­plines,” it’s com­mon­ly prac­ticed as 8D. Learn how to form teams, con­tain prob­lems, scope projects, iden­ti­fy root caus­es, imple­ment cor­rec­tive actions and pre­ven­tive mea­sures, and rec­og­nize the team’s success.

In the Six Sig­ma DMA­IC approach we iden­ti­fy, eval­u­ate, and opti­mize the crit­i­cal inputs, or Xs, that sig­nif­i­cant­ly influ­ence our pri­ma­ry met­ric and over­all process per­for­mance. Learn how to use tools like process map­ping, the Cause & Effect Matrix, and FMEA to iden­ti­fy process inputs and nar­row them down to the crit­i­cal Xs.