Patterns: What Are They? Part 1

Literally patterns are defined as a combination of qualities, items, objects, behaviors, or other which form a consistent or characteristic arrangement. We observe dozens of daily patterns every day. Examples are teen behavior, the movement of the earth around the sun, the blooming of flowers and hundreds more.

"A combination of qualities, items, objects, behaviors, or other which form a consistent or characteristic arrangement"

Patterns are not new to man-kind, the Creator has informed us about patterns in the Quran and other earlier revelations. Recently system developers started to analyze the concept of patterns, pioneered by Christopher Alexander in his book, " A Timeless Way of Building", he defines a pattern as a three-part rule, expressing a relation between a certain context, a problem and a solution.

Patterns are a first step in understanding root causes of the problem within a specific context. It also allows the development of architectural and solution building blocks.

"In systems realization, patterns express a relationship between a problem and its solution within a given context"

For a software designer some pattern examples are: Message Translators, Process Managers, and Proxies to mention a few. A process manager pattern for example will comprise of a sensor to detect an incoming message which initializes the process manager. The process manager in turn executes a set of rules stored in the process manager memory implementing the processing rules, and detects the subsequent steps through a status feedback detector/analyzer.

Considering a different domain, take the example of a group that does youth trips, examples of patterns could be: Headcount Process, Critical Supplies, Meal Managers, Activity Itinerary Manager and many more. An Activity Itinerary Manager pattern will comprise of an input process to capture developmental and coaching themes/needs of the youth trip, a search process to match the theme to a group of activities that meet the objectives of the theme from a repository of activities, the selected lists of activities could then pass by the activity processor which breaks down the activity into steps, each assigned to resources, a lead, a cost structure, risk factor and plan of action. Feedback from the actual deployed activity is then finally fed back into the activity manager for updates to the activity definitions and other relevant areas, within quality expectations.

_____________To Read More _______________

(1) Hohpe and Woolfe, "Enterprise Integration Patterns: Designing, Building and Deploying Messaging Solutions", Addison Wesley signature Series, 2004.

(2) Haskins and Raveh, "Introduction to Patterns Through Writing Systems Engineering Patterns", 16th Annual International Symposium, 2006.

(3) Gross and Yu, "From Non-Functional Requirements to Design Through Patterns".

Six Sigma's DMAIC vs. Design for Six Sigma's IDOV / DMADV

Six sigma's success is mainly contributed to its DMAIC model. The DMAIC model focuses on defining an issue with an existing process or product, measuring the deviation from the expected behavior, analyzing the data for insights to root causes, improve the current process or product through minimization of deviations or improvements and finally controlling the system to sustain the gains achieved.

IDOV and DMADV which are two design-based models slightly differ from DMAIC as follows.

Objective and Approach: DMAIC views the current process or product as correct and economical, but needs to minimize some gap leading to inefficiencies. IDOV / DMADV view the current process or product as in need of redesign or design change to achieve customer satisfaction.

Process Capability: DMAIC views current process as capable of satisfying customer needs, whereas IDOV / DMADV views current processes as a candidate for improved yield regardless of volume and complexity.

Design: DMAIC views the current design as satisfactory for the client's needs, whereas IDOV / DMADV views the need to consider various drivers for design, such as cost, manufacturing, producibility, maintainability, robustness, usability, efficiency, security, agility, compliance and testability.

Flexibility: DMAIC assumes that the current design and processes are flexible to meet customer demands and needs, whereas IDOV / DMADV highly considers potential customer demands and forecasts newly developed needs

Validation: IDOV and DMADV both consider validation and verification of the outcomes of a design or process in meeting objectives.

To read more:
[1] Chrisitian Madu, "The House of Quality in a Minute", Chi Publishers, 2006.
[2] Rod Munro, "The Certified Six Sigma Green Belt Handbook", Quality Press, 2008.

3 Practices to Becoming an Agile and Lean Enterprise

About a month ago I was approached by a client asking to give advice regarding some organizational changes and major layoffs they were embarking upon. The client sent me an email with about six or seven different options to pursue to cut on their operating expenses.

I noticed that all the options were focused on eliminating employees. My question to myself, if eliminating employees seems the favorable and only approach that comes to mind, why were these employees present in the first place? There must have been a need for them in the organization, otherwise why were they hired in the first place? What operations will suffer when these folks are let go?

Many businesses fail to realize that a lot could be done way in advance to avoid shedding off employees. Letting go of your most valuable resources, your staff, should be the last thing to do, before closing shop. I list ten actions that can be done months before the crunch becomes severe to be forced to shut down.

1. Reduce non-valuable activities, a.k.a trivial work.
Non-valuable activities are those which your customers are not willing to pay for, they do not change the form or function of the product or service they are interested in. An example is rework, due to defects in a process used to produce the end product or service. Other examples are wait-time during the process. For example if you order a book online, you are willing to pay for the book to be shipped from the publisher to your location, you are interested in how much time it takes the book to wait at each regional hub for its next pickup. If there is a cost to the shipper to pay for the time spent at each shipping hub it will increase the cost of your book shipment and will not provide you any value. The only value you realize is the book being shipped to you as fast as possible. Instead of focusing on accelerating valuable processes, one should first eliminate non-valuable processes, as the effort of work will probably be less and the outcomes more effective.

Muslims also know non-valuable activities as "Lagow", an Arabic term mentioned in the Quran in several places, as deeds that have no benefit, or are a mere waste of time. These are deeds that keeps one's focus away from his purpose in life, which is the success in the hereafter.

Before improving valuable processes, it is important to get rid of as many non-valuable processes as possible. In the case of my client, they were focussing on products that yielded low profit margins, were difficult to market, promote and sell, and required large staffing and overhead.

2. Reduce waste
Waste is all around us in our activities. It is imperative that before we develop new processes to improve deficiencies we actually reduce waste as much as possible. Waste could be due to over production, excessive steps due to complexities of processes, queuing and idle time, defect correction and rework, poor organizational and planning processes, lack of controls and creativity stagnation.

Through some auditing and analysis it was found that my client had opportunities to reduce expenses of photocopying, utilities, rent and other non-valuable activities prior to let go of employees who were driving valuable activities for the end user.

3. Use facilitation to reduce cycle time and enhance efficiencies
A facilitator who is neutral to the different entities in your enterprise, and who has no authority or decision-making control, can bring tremendous value to your organization. The facilitator can apply collaboration concepts, systemic approaches and architectural thinking to decompose problems, address impacts in all areas of interest, and ensure a collaborative environment exists where the various team members can share ideas, brainstorm and foster creativity.

Controling Many Small Changes: Taming the Culprit for Most Accidents or Failures

As changes creep upon our processes, methods and systems; we accumulate little bits of inefficiencies which eventually become large and cripple the system.

Six Sigma helps ensure continuous improvement, through the use of data collection to analyze and understand how a process works or changes. The focus in this case is to reduce variations to processes or design ultimately leading to less defects. Common approaches to identify these variations or deltas is through the voice of the customer (VOC), quality function deployment (QFD), failure mode effect analysis (FMEA) among many others.

Lean improvement approaches can also help address change creep consequences. The focus in this case is more on waste minimization and elimination. Common approaches are value stream analysis, the five S, Kaizen events, just in time, work standardization and error proofing.

__________________________

Picture: The Nile Delta in Northern Egypt. Image depicts delta erosion.

Feedback: A Core Component of a Successful Process

Processes comprise of a functions which operate on inputs to produce outputs. Data associated with the inputs and outputs can be used to further enhance the process quality. The data related to the process outputs could include important elements that when fed back to the input within a solid analytical framework could offer valuable insights and impacts on the process improvement.

Several key steps in designing the appropriate data collection, analysis and feedback system are summarized below:

1. Decide where to collect the data
Data can be collected at the input, various processing points, decision-points, at the output or any combination of all of these locations.

2. Decide what measurement systems to use
Various techniques exist to collect and measure data. Decisions need to be made on best statistical approaches, summarization techniques, collection methods and the levels of accuracy and precision needed.

3. Decide how to analyze the data
Approaches of data analysis need to be determined. It is important to determine cause and effects, root causes, correlation, regression dependencies and other relationships among the various variables and factors.

4. Decide on how to use the information resulting from the data analysis
Decisions related to the application of the information learned is important. For example will information be applied in real-time, or after process reengineering is brainstormed.