Design of Experiment (DoE)

Design of Experiment (DoE) is a structured, organized method that is used to determine the relationship between the different factors (Xs) affecting a process and the output of that process (Y). This method was first developed in the 1920s and 1930, by Sir Ronald A. Fisher, the renowned mathematician and geneticist.

Design of Experiment involves designing a set of ten to twenty experiments, in which all relevant factors are varied systematically. When the results of these experiments are analyzed, they help to identify optimal conditions, the factors that most influence the results, and those that do not, as well as details such as the existence of interactions and synergies between factors.

DoE methods require well-structured data matrices. When applied to a well-structured matrix, analysis of variance delivers accurate results, even when the matrix that is analyzed is quite small. Today, Fisher's methods of design and analysis are international standards in business and applied science.

Experimental design is a strategy to gather empirical knowledge, i.e. knowledge based on the analysis of experimental data and not on theoretical models. It can be applied whenever you intend to investigate a phenomenon in order to gain understanding or improve performance.

Building a design means, carefully choosing a small number of experiments that are to be performed under controlled conditions. There are four interrelated steps in building a design:

   1.     Define an objective to the investigation, e.g. better understand or sort out important variables or find optimum.
   2.     Define the variables that will be controlled during the experiment (design variables), and their levels or ranges of variation.
   3.     Define the variables that will be measured to describe the outcome of the experimental runs (response variables), and examine their precision.
   4.     Among the available standard designs, choose the one that is compatible with the objective, number of design variables and precision of measurements, and has a reasonable cost.

Standard designs are well-known classes of experimental designs. They can be generated automatically as soon as you have decided on the objective, the number and nature of design variables, the nature of the responses and the number of experimental runs you can afford. Generating such a design will provide you with a list of all experiments you must perform, to gather enough information for your purposes.

Design of Experiments (DoE) is widely used in research and development, where a large proportion of the resources go towards solving optimization problems. The key to minimizing optimization costs is to conduct as few experiments as possible. DoE requires only a small set of experiments and thus helps to reduce costs

SPC

Statistical Process Control (SPC) can be applied to software development processes. A process has one or more outputs, as depicted in the figure below. These outputs, in turn, have measurable attributes. SPC is based on the idea that these attributes have two sources of variation: natural (also known as common) and assignable (also known as special) causes. If the observed variability of the attributes of a process is within the range of variability from natural causes, the process is said to be under statistical control. The practitioner of SPC tracks the variability of the process to be controlled. When that variability exceeds the range to be expected from natural causes, one then identifies and corrects assignable causes.

SPC is a powerful tool to optimize the amount of information needed for use in making management decisions. Statistical techniques provide an understanding of the business baselines, insights for process improvements, communication of value and results of processes, and active and visible involvement. SPC provides real time analysis to establish controllable process baselines; learn, set, and dynamically improve process capabilities; and focus business on areas needing improvement. SPC moves away from opinion-based decision making.

These benefits of SPC cannot be obtained immediately by all organizations. SPC requires defined processes and a discipline of following them. It requires a climate in which personnel are not punished when problems are detected, and strong management commitment.

Definition of Six Sigma

There is a saying with regards to quality that if it is not 100%, then the quality of the product or service is bad. Six Sigma is a set of practices which tries to achieve just that, near perfection in Quality. If I would tell you that the Quality standards as per Six Sigma standards is 99.99998%, you would infer directly that it is almost near perfection. And that 99.99998% translates to 3.4 Defects per Million opportunities, which means 3.4 errors for one million transactions. That is a huge representative for perfection.

Some myth-busters about Six Sigma

•Six Sigma is very subjective in its approach - It is completely the inverse of it. Six Sigma is fuelled by data all the way with very little room for text. Even the performance improvement measures which Six Sigma advocates through its DMAIC model is data-driven.


•Six Sigma can only measure quality only- Though Six Sigma is set of quality practices, the practices are developed and targeted in such a way that a quantitative metric could also be measured very easily. The Six Sigma process does not stop with measurement of data. It goes some more steps ahead and analysis is performed, improvement measures are suggested and then the improvement measures-driven data is measured. All these steps come together to reengineer a process to ensure that the objectives of the business are always met.


•One can only improve on existing processes with Six Sigma - If one had to define Six Sigma in industry terms, it would be something like this "It is a set of practices which reengineers processes to ensure that customer satisfaction levels are met always. In case, process does not existed, process can be engineered to ensure that customer satisfaction levels are met." Essentially, the DMADV model of Six Sigma allows you to design and validate new processes which could be possible factors to impact customer satisfaction.


•Six Sigma is all about Green Belt/Black Belt/Yellow Belt - It is exactly the opposite. Green/Black/Yellow belt certified professionals who are trained in Six Sigma. But the success of a Six Sigma process improvement measure can be successful only due to two factors - The strength of the process and the ability and the cooperation of the employees to execute the project.
Six Sigma with its tools and statistical controls has become a synonymous name with companies who wish to establish strong business processes which will deliver results with the minimum variations. Though, it initially started of being a set of practices for the manufacturing sector, it is catching up speed in other industries as well.