Monday, August 3, 2015

There Is No Substitute For Good Engineering

Success, be it in a project or any other human endeavor, never happens spontaneously. It is the result, not merely of effort, but also of deliberation, of thought--i.e., of design.  As the above litany of statistics demonstrates, a lack of design equates to a lack of success.  My assertion is that the inverse also holds true--that quality design equates to success.  As a direct consequence, it behooves us to contemplate what makes for a "quality" design in virtually all fields of human endeavor.

The body of thought arising from the contemplation of quality design may be fairly considered as the discipline of engineering.  The close connection between each of the terms, as well as the concepts, is easily intuited by consulting any dictionary (my personal preference is Merriam-Webster) and noting that "engineering" is defined in terms of "design": "the work of designing and creating large structures (such as roads and bridges) or new products or systems by using scientific methods." Although we habitually use the term "engineering" in relation to large construction projects, within the basic definition one can easily see how engineering concepts can be brought to bear in almost any field of human endeavor.  "Engineering", to my mind, is the science of design.

Understanding any science begins with an understanding of its first principles. To understand, then, the science of good design we must first examine the principles and the process of design itself--and there is indeed a logical conceptual process that informs all design efforts (and the quality of our design results hinges on the extent to which we leverage that process in all our design activities.

An excellent depiction of the design process comes to us courtesy of the Massachusetts Institute of Technology's Open CourseWare initiative, which makes a variety of classroom lectures and supporting materials freely available online.  In particular, I make reference here to the 10-step design process outlined in ESD.051J Engineering Innovation and Design.  As presented in that lecture series, the ten steps in the fundamental design process are as follows:
Identify Needs
What problem are we attempting to solve? What are our goals and objectives?

Information Phase
Do any solutions for the problem already exist?  What are they? What are their strengths and weaknesses?

Stakeholder Phase
Who desires which specific goal or objective? Why?

Planning/Operational Research
What's realistic? What limits us?

Hazard Analyses
What's safe? What can go wrong?

Specifications
What's required? How do we define success?

Creative Design
Ideation

Conceptual Design
Potential solutions

Prototype Design
Create a version of the preferred design

Verification
Does it work? If not, redesign
Note: I do not intend an in-depth exploration of these phases here. I highly recommend visiting the MIT Open Courseware site for a more thorough presentation of this process.
What is noteworthy about this process is that it is not tied to any particular technological or scientific paradigm.  Indeed, it is not tied to any particular discipline at all.  One could apply this process not only to the development of industrial products and technologies, but also to project plans, marketing plans, and even the creation of art (this is not as far-fetched as it might seem--several of the world's foremost industrial designers, such as Dieter Rams and Apple's Chief Design Officer Sir Jonathan Ive began their studies within the world of art, and Dieter Rams' designs have been gathered into a traveling exhibit Less and More, which has been displayed at art museums around the world).

This process works because it is comprehensive. It demands that one reach outside of his or her own thoughts and ideas, and specifically engage with others, solicit input and ultimately feedback from others. Throughout, the design process explicitly seeks connection to the reality surrounding the effort, and, in an iterative fashion, requires constant validation of effort against that reality.

With so much attention devoted to design, why do I speak of engineering in the title of this posting? The answer is simply this: Conceptual consideration of a design process, or of any process, itself yields no tangible results.  That conceptual consideration must be given form and function through practical usage. That practical usage is the discipline of engineering.

In 2012, Dr. Benoit Hardy-Vallee, writing for the Gallup organization, reported on a study published in the Harvard Business Journal that found, out of 1,471 IT projects, average cost overruns of 27%, with one in six projects having cost overruns of over 200% and schedule overruns of almost 70%.  Dr. Hardy-Vallee categorized project failure causes into three broad categories:
  • technical (technology developed, project management techniques)
  • individual (project leadership, scope management, communication)
  • stakeholder (user involvement, executive buy-in, goal specificity)
Dr. Hardy-Vallee concluded that project management failure was largely result of inadequate "emotional" content. His assertion was that project management current literature was rich on process controls and methodologies, but severely lacking in ways to engage various project stakeholders at an emotive and psychological level:
The problem with a single-minded focus on processes and methodologies is that once people are given procedures to follow, compliance replaces results. Everybody is concerned about how to do the job, not about the outcome if the job is done well.
However, when one considers Dr.  Hardy-Vallee's causes of failure against the backdrop of the fundamental process of engineering design outlined above, one can match the causes of failure to particular phases of the design process (and sometimes to more than one phase of the design process), and so a different conclusion becomes possible--that projects fail from poor design and from poor implementation of a design. Projects fail because of a lack of good engineering. 

In an engineering paradigm, processes that do not produce good results are not good processes; they should be altered and adapted until they do produce good results. Likewise, methodologies that do not yield successful outcomes are not successful methodologies, and should be revised so that successful outcomes are obtained. Because of the iterative nature of the design process, "compliance" in the engineering paradigm becomes the pathway to results, and thus is the complement to those results and never its substitute. (The "engineering" response to Dr. Hardy-Vallee's emphasis on a need for "emotional" content within project management would be that by definition, effective stakeholder analysis phase of design engages all project participants at that emotive and psychological level).

No matter the project, whether building a bridge, designing a video game, or marketing a new product, success comes from the careful and deliberate application of the design process. Success comes from good engineering.

No comments :

Post a Comment