Development of a Composite Model for Six Sigma and Data Analytics for Organizational Evolution

Six Sigma took its birth in providing a structured approach to problem-solving by means of defect reduction in Organisations. The focus subsequently moved to waste reduction in processes without compromising on the level of defects and was termed as Lean Six Sigma approach (LSS). Over time, the attention again shifted to the causes which gave rise to defects and waste and ways to prevent it from birth. This gave rise to another domain-intensive approach called Design for Six Sigma (DFSS) which was essentially a design and development intensive approach. All three approaches of Six Sigma together provide solutions to various types of problems and challenges that organizations are encountering. However, in today’s VUCA world characterized by Volatility, Uncertainty, Complexity, and Ambiguity; many times, the problem domain itself is fuzzy. The discipline of Data Analytics helps to clear this cloud of confusion and helps to reach a conclusion about the problem domain. These two methodologies are complementary in nature and this paper is proposing a composite model on how they can work together to promote business evolution. Not much literature is available on the subject and this model is intended to fill this gap. The paper focuses more on the business aspect of the model, and is described in layman terms, and avoids technical jargon.

Keywords

DMAIC, LSS, DFSS, DMADV, DATA ANALYTICS

Background

To appreciate this paper, it is important to briefly explore the term Sigma Level which is used frequently in this paper and forms the basis of the term Six Sigma. Let us just for the time being recognize the fact that an increase in Sigma Level reduces no defects per million instances or units or opportunities. So, as we move from one Sigma to two Sigma, the number of defects per million reduces. Hence higher the Sigma level, the lower will be the defects per million. It is important to note that the relationship is not linear and depends on what we call a normal distribution, popularly known as a bell curve. This happens due to a reduction in the variability in the process which improves its capability of consistently hitting the target. To get an idea of how defect level varies with Sigma Level, please see the following table. Figures are rounded appropriately for easy readability:

The Six Sigma approach to problem-solving for defect reduction has been present since 1979. The concept was born and nurtured in Motorola and thereafter was adopted by such iconic organisations like Allied Signals and General Electric (GE). The benefits realised by these organisations helped Six Sigma to gain recognition and get established as preferred problem solving approach for defect reduction. The rise of Six Sigma is well documented and many books have been written by practitioners including Dr. Michael Harry. Dr. Harry focussed on approach R-DMAIC-SI which consists of eight phases: Recognise, Define, Measure, Analyse, Improve, Control, Standardise and Integrate. Whereas the phases DMAIC operate on individual problems, rest of the three phases R and SI works at organisational level. In technical terms, Six sigma is methodology for progressively, aggressively and incessantly keep on reducing variability to reach to defect level below 3.4 parts per million.Six Sigma directly impacts the profit of the Organisation and positively impacts other soft parameters like Customer Satisfaction, Employee Morale, and Company Brand image.

Lean Six Sigma started around 2002 to reduce waste in complex processes and at the same time does not impact the defect levels. The need for LSS was felt because the Six Sigma DMAIC process was found to have limited application in waste reduction. Many organizations like Toyota very successfully implemented LSS and the fact that it accrued quick benefits in Cash Flow made it very popular across the world. Six Sigma (DMAIC) along with LSS typically have a positive impact on the bottom line of the Organisations.

Reducing defect rate and waste below a certain level becomes very difficult both technologically difficult and costly. Many times, it is not possible due to constraints and limitations inherent in the system. In technical terms, it is very difficult to improve beyond the four-sigma level due to various limitations. The thought process very naturally goes to “Why defects and waste are present in the System? Is there any way to design a System so that defects and waste are not present from birth?” This exploration of this question slowly gave rise to a discipline known as Design for Six Sigma. DFSS ensures that the created outfit in terms of new product or process meets the requirements of the Six Sigma level from the launch. DFSS primarily focuses on design and development across all the life stages of the Product Development Life Cycle so that they are insulated from defects and waste throughout its life. DFSS typically has a positive impact on the top line of an organization with the protected bottom line.

The genesis of Data Analytics lies in the field of Statistics and Mathematics and can be considered as old as the process of collection and interpretation of data itself. The process of collecting census data for planning and managing affairs of the state has been happening across the world and is believed to predate Egyptian civilization. Naturally, these data were required to be analyzed and interpreted. Analysis of data used to be time taking and the complicated process took years of painstaking work to arrive at the results. The advent of computers and advances in computing technology provided a major thrust to the development and application of Data Analytics in a variety of Areas. The modern era of big data and Cloud has resulted in the true power of Data Analytics being unleashed with almost all companies, big and small adopting it for their business.

In simple terms, Data Analytics is the discovery, interpretation, and communication of meaningful (often hidden and unexpected) patterns in data. Finding such patterns is commonly referred to as Knowledge Discovery using Data (KDD). KDD provides clarity to a situation or answers a question, moving the status from confusion to conclusion, thus clarifying the problem domain where solutioning needs to take place.

It should be noted here that Data Analytics is still an evolving field and no universally accepted structure exists. Here we will consider the following 7 step process:

Objectives

To develop a composite model of Six Sigma (consisting of its three Avatars - DMAIC, DFSS, and LSS) and Data Analytics, and to connect it to the Organisation’s growth aspirations.

Proposed model

At the highest level, the Proposed Model is essentially a two-step process: Knowledge or Insight discovery and Solutioning on the issues and opportunities identified. The model is depicted in the diagram below:

The knowledge/ Insight discovery process lies both in the domain of Six Sigma and Data Analytics. In the domain of Data Analytics, KDD i.e., Knowledge Discovery using Data is performed. The seven-step process of Data Analytics is used for this purpose. Many of the time, the internal issues, problems, and challenges are well known hence Data-driven knowledge/Insight Discovery operates more in relation to the External Ecosystem or in VUCA like situations within the Organisation. It is also applied to situations where massive datasets are generated, and it is not possible to manually perform any Analysis.

In the domain of Six Sigma, the Knowledge/Insight Discovery process is applied by Organisation to understand itself better. On the one hand, this understanding comes from periodically exploring, examining, and evaluating its Vision, Aspirations, Strategies, and Strategic Business Objectives. This exercise is usually performed by the apex and senior management. On the other hand, the Opportunities and Issues need to be discovered by exposing the hidden factory within the Organisation by the Survey of Departments and Stakeholders.

Knowledge/Insight Discovery using Data Analytics and Six Sigma thus result in the identification of Business expansion and improvement opportunities.

Solutioning scope lies more internal to the company within the area of influence. Business Expansion and Improvement Opportunities identified during Knowledge/ Insight Discovery are further refined to Projects which can be the solution to get the Business Results. The Projects can be executed using the appropriate Six Sigma approach i.e. DMAIC, LSS, or DFSS, or a combination of these.

It is important to note that the steps leading up to Project identification and prioritization are done at the Organisational level. The Solutioning phase thereafter is executed at the Project level. In a complex situation, Data Analytics tools are applied during the solution phase while defining data models in the project Analyse phase. It helps to establish exploratory and causal relationships between critical variables. The overall model is depicted in the following figure.

Once the projects have ended and Business goals are achieved, the learnings are incorporated in the integrated Business Management System (BMS) as standard processes. Input is also received from Data Analytics in the form of learnings while working on Data and moving from confusion to conclusion. Once the learnings are Standardised, the collective learnings in the form of integrated BMS are integrated into the culture and gene of the Organisation using Performance Management System. Periodic performance assessment and Corrective action are undertaken to check any degradation to the Standardised and Integrated System. Note that these entire phases of Standardisation, Integration and Performance Assessment & Corrective Action help in taking the improved state to the Organisation’s gene hence this phase is called Genetising.

It may be observed that the proposed model consists of layers of models. It would not be wrong to say that it is a model on top of models. It can be seen pictorially depicted as below:

Illustrations of the proposed model with a Case Study

Conclusion

The Composite Model for Six Sigma and Data Analytics provides a way to achieve Organisational Evolution. It starts with Knowledge Discovery working on Vision and Strategy of Business and working its way down the organization to discover the hidden factory of Business expansion and improvement opportunities; Analytics of Data to discover hidden knowledge and patterns plays a very important role in this exercise. This is performed in the context of both External and Internal ecosystems. The identified opportunities are solutions to get the desired business result. The composite learning is then genetised by Standardisation, Integration, and periodic Performance Assessment & Corrective Action.

Previous