In the Beginning

We have called the scope of work in “making sure the city is worth the siege,” The Front End for fifty years. Too many times in our large project experience we diligently plugged away only to find out, near the end when the money was gone, the basic assumptions of the project were grossly misaligned with reality. In time we learned that all big projects are conceived and launched by authoritarian folks, bereft of trustworthy information, who don’t have their act together.

As one example, during the kickoff meeting of the Big Dig in Boston, in 1990, everyone in attendance knew from mutual GYAT assessment that it was going to be a disaster. From that event forward, all project activity was defensive. No one claimed responsibility for project success. Litigation went on for decades.

The first publication about Plan B (1986) and every book since has featured The Front End (TFE) and the law of optimality that shapes it. The front end is where the work flow of Plan B a priori begins along with ground truth acquisition. TFE, mostly about system statics and uncertainty identification ends with the handoff to dynamics where uncertainty is enumerated and systematically eliminated by hard system performance data on trustworthy facsimiles.

Those chained to business as usual despise “The Front End.” With no role for opinion, they know all the errors being found are going to make them look bad and social status trumps project attainment every time. The truth is The Front End makes everyone look bad, including those of us who execute its procedures. Finding out how many of your innate perceptions and intuitions about the project were wrong is a cold shower.

Once TFE is completed, however, everyone is delighted that The Front End work was performed. Starting with the truth, dodging the GIGO bullet, is comforting. With people who have demonstrated having their act together, they know the project is going to be a success and want to be part of it. Its transparency is proudly displayed on workroom walls.

Whenever you are tasked to project work, novel and temporary, there are some start-right assumptions. Whatever the given specifications, you can be sure they were prepared when project knowledge was sparse, missing, and unreliable. Further, the people who made the initial choices, entitled authoritarians, were unlikely to have Actionable Quality Information (AQI) or their act together.

Accordingly, respecting the law of optimality, you can assume the information provided at project’s outset is too unreliable to use as a foundation for going forward. Validating the project scope by acquired ground truth becomes the first order of business. There is no way to build a process that starts where you are and leads to the goal without a GPS of the goal.  You must make good choices at each step before going on to the next step and there are benchmarks for checking that that is the case. In the end, system performance answers all questions about the GYAT choice-making process.

In validating the primary function, you visit the sources of the assumptions that management used to choose the ultimate objective. Experience shows you will get an education about the confirmation bias. Those with ground truth will be happy to discuss the project with someone who seeks the truth.

Every effort to check out the source of the project goal will pay dividends. The exercise is to start right, not embarrass the head shed. This knowledge is your sanctuary from administrative meddling to protect their bad choices.


Scope and task progression of The Front End (TFE)

  1. Goal validation: ground truthing. Sources of choices made so far
    1. Starting assumptions of as-received project (POSIWID)
    2. Production Goal  primary function, generalized functions (Rasmussen)
    3. Design Basis Events (DBE)
  2. Franceschi Fitting ground truth
  3. Stakeholders
  4. Forecast project context, without intervention
    1. Legitimate responsibility for satisfactory results
    2. Prevailing project morale
    3. Husbandry provisions
  5. Reliability engineering
  6. What done looks like – physically
    1. Design basis flow diagram
  7. Regulatory climate, licensing and permits
  8. Administration:
    1. Wrap-up documentation of TFE work: design basis, flow, assessment, projections, and recommendations
  9. Handoff to dynamics, Premortem dynamic uncertainties.


  • There can be no losers within the Plan B envelope.
  • All choices are made by using only GIGO-free AQI (actionable-quality information)
  • Everything in the flow path has live implementations available for audit and is demonstrable in your own shop.
  • Everything is proudly transparent and falsifiable.

Ground truth acquisition

In the case of a project engagement, you start with validating the mission purpose. To validate requires obtaining ground truth first hand, reconnaissance. Read, but ignore official sources. Ground truth comes from personal, one on one, direct contact with the MitMs involved in the proceedings. It takes time and effort, but if you are GYAT, the MitMs will give you an earful.

If you detect deliberate bias in the formulation of the project scope, you can be sure the project is doomed. Bias, distortion of reality, will show up everywhere. It is GIGO contamination writ large.

If the ground truth on the project goal statement and prime function is validated, proceed down the Rasmussen hierarchy, to generalized functions to the Franceschi Fitting (FF) level. This is the critical translation between abstraction/authority and materiality. The generalized functions themselves point nowhere to the material world.

The creation of the physical functions that constitute project success is pure invention that, of necessity, must be done by someone who knows both the fictional world of the authoritarians and the real world of the producers, tangible system performance and goal realization.

Because it is 100% creativity, a single mind is involved. To interpret mission success in terms of physical functions, he takes a generalized function as input and invents a set of physical functions he thinks will deliver the objective. Once defined, selecting the tangibles to produce that function fits routine engineering, using precedent and experience, expressed in flow diagrams and part numbers. The validation task is to ground truth the genius who custom-invented and expressed the FF.

Once the part numbers have been selected, it is possible to create what done looks like physically. What used to be an artist’s rendition has become a packaged software chore.



H.S. Dennison, 1922

“The primary data of any project in organization engineering are the special characteristics of just those human beings by whom it is to be manned. Proposals for representative forms of government, for liberty of interpretation, for measures of military discipline, or for functionalized factory management cannot be judged in abstraction. Their discussion is significant only as applied to men and women of known characteristics.

The men and women members of an organization, to whom its incentives and its rules apply, vary widely among themselves in physical powers, in mental development, and in their responses to emotional stimuli. Organization engineering has first to discover within any given group the prevailing conditions and develop its forms of organization and general operating measures so as to accommodate itself to them.

It will also, where it is possible by selection and training, make closer adaptation of its members to the purposes of the organization and to each other. The ultimate in organization would require that the incentives, regulations, and personal contacts of each member should be such as to allow him to develop and put to most valuable use all of his powers. Since no two members are exactly alike there would be, theoretically, separate provisions for each. But all members, though different, are compounded of quite similar elements; their differences are in large part differences in the proportions in which these elements are represented.

Hence, some reasonably uniform general measures can usually be devised to appeal to, to satisfy, and to control the more widely prevalent elements of human nature. An important practical task of the organization engineer lies in so grouping the personnel and devising the general rules that the largest possible number are effectively provided for; at the same time he must be ready to deal with individual cases wherever the unsuitability of general rules causes losses which outweigh their gains in convenience and cost of administration.”

Operating environment

The task paralleling the Franceschi Fitting innovation is to get real with the likely project operating environment. It’s basically about trust. If Plan A, no trust, no responsibility-taking. All work is defensive. In this case, add a task to the pile to find a way to run the Front Line Leader Program (FLLP). It is the difference between certain project failure and project success and you know which is which before the project funds are exhausted.

Determining if Plan A or Plan B context, is the easiest task in the project. You can’t get it wrong even if you try.


In Plan B, producer world, no one cares about your ethics or morals. What does count is your productivity towards the common goal. Producer world can’t understand any other definition of “responsibility” than process results. When you volunteer to take outcome responsibility in producer world, you are given the autonomy necessary to meet your pledge. The rest is up to you. That is, exactly, a square deal.

There are sure-fire indicators that no one is responsible for obtaining satisfactory project outcomes.

  • Authorities pronounce that everyone is responsible for results
  • Authorities claim to be responsible for outcomes
  • No one claims responsibility for outcomes
  • Defensive routines, undiscussables
  • Groupthink is the operating ideology

The only valid responsibility-taking for successful outcomes is someone who has GYAT and knows the Plan B standard of care process.

Anyone who claims responsibility for the project deliverable who is not GYAT, trustworthy, is immediately sensed as an imposter on the spot at the project kickoff meeting. Any doubts about this is answered by any Congressional hearing on corporate malfeasance. The common excuse of the CEO imposters for the wreckage as questioned is “I was too remote from the corruption to know about the corruption going on.” No one is to blame. In Plan B, the legitimate taker of responsibility is public knowledge.

In large projects, applications have to be filed with government agencies that designate responsibility. The professional responsible under tort law is called the “engineer of record.” There is no responsibility of management to deliver a successful project, because it can’t


Design Basis Events

Design Basis Events are those foreseeable operating disturbances that judgment and opinion deem worthy of prevention, avoidance, and survival in the system design. This list of events usually includes earthquakes, tsunamis, fire, pipe ruptures, etc. The list expands along with growth in project knowledge.


Authoritarianism doesn’t care a whit about stakeholder satisfaction and never did. The entitled consumers of the upper hierarchy do not compile and publish a stakeholders list with a definition of the stakes of each. Any conflicts between the actions being taken and stakeholder interests are ignored. The only backstop for stakeholder abuse is tort law – after the damage has been incurred.

Plan B policy holds there are to be no losers, including stakeholders. The no-loser dictum solves the problem of getting their stakes included in project navigation considerations. In Plan B, deliberate effort is directed to identify stakeholder interests and their expectations. In most situations, success of the project takes care of the stakeholder obligation.

Reliability Engineering

System reliability requirements must be explicit and accommodated before the configuration of the process takes place. The discipline that covers this, reliability engineering, uses equipment failure data and mathematical physics to define appropriate system reliability actions. There are many options available to correct deficiencies in reliability including redundancy and preventative maintenance. Plan A has to be driven to consider reliability. Plan B includes it under the instinct of workmanship.

Regulatory climate

In Plan B, since the producers have their act together, the regulator agencies, all Plan As, are handled in a preemptive manner. You have the design basis, incontrovertible, transparency, and demonstrability. They are shown the impact of Plan B on their agency goal. They are offered training to help their clients understand the significance of Plan B. When you have Plan B, your strategy with other institutions is offense, not defense.

Wrap-up Report

The purpose of documenting completion of the front end phase is to make a GIGO-free start of the dynamics phase. Transparency is essential. Not everyone who worked the front end will be working the uncertainty elimination portion of the project. The informal report is available as a confidence-builder for the stakeholders.

The documentation includes work details, observations, evidence, and forecasts of project progress going forward.

The world of shapes, lines, curves, and solids is as varied as the world of numbers, and it is only our long-satisfied possession of Euclidean geometry that offers us the impression, or the illusion, that it has, that world, already been encompassed in a manageable intellectual structure. The lineaments of that structure are well known: as in the rest of life, something is given and something is gotten; but the logic behind those lineaments is apt to pass unnoticed, and it is the logic that controls the system. David Berlinski

Views: 129