Presentation: Four Best Practices -Large Scale Projects

Following the attack at Pearl Harbor in 1941, George Marshall asked Dwight Eisenhower to come up with a war strategy.   Europe was being over-run by German and England had barely held off the German air force in 1940.  Japan's attack on the US now brought the US into war and the US needed allies.  Large project tend to generate a lot of plans but often basic principles are overlooked.  Where is equipment going?  Is there enough space?  Is the sequence of events correct?     One project had over 160 Automation plans without any type of overarching plan that answered basic questions.  And, sure enough, one of the units was to startup before the associated control room was to be finished.   How many plans did Eisenhower generate that established the US Strategy in World War II that mobilized more people and equipment in world History?   It was nowhere near 160.  It was one one plan.  And that plan was 7 pages that established basic strategies.     One strategy was to address Germany first rather than Japan.  At the time,  the US wanted to come after Japan due to Pearl Harbor.  But Eisenhower reasoned that Germany was on the verge of taking over Middle East oil and would be much more difficult if not impossible to fight.  In addition, with England on the verge of collapsing,  the US would have a difficult time waging war from the US without any type of local Allied foothold in Europe.  A D-day type invasion across the Atlantic would have been impossible.     The strategies that Eisenhower outlined turned out to be correct and led to Allied victory.  If WWII only needed a 7 page plan, why should projects try to characterize high level planning across 160 documents.  Projects with that many plans need a high level strategy that all understand and agree to.  It should align with the high level objectives of the project.  It is surprising to see how frequently this high level plan is missing and the project team is not clear on the objectives.  Those projects often fail.    John F. Kennedy frequently consulted with Eisenhower on strategy despite both were in opposing political parties.  And like WWII, the successful Apollo Moon Landing that Kennedy established became one of the most successful endeavors in history.  The original objective of "landing a man on the moon before the decade is out" was accomplished on July 20, 1969 and only a few months from the target date.   And again, this program started with an 8 page strategy list of questions similar to how Eisenhower presented his strategy.  WWII and landing on the Moon are much like large, complex projects with huge unknown risks.    These large, complex projects can be as successful by following some of the same strategy principles. Once the strategy is set,  execution needs to maintain adherence to the objectives that frequently become lost.  What's something exciting your business offers? Say it here.

"If someone is at the site managing the work,  make sure they are you're best".   Experience and insight are frequently overlooked as a quality in people.  Psychologist Gary Klein conducted studies in 1985 to determine how fire fighters made critical life and death decisions and came across the intuition process used by many professional including not just firefighters but military, medical and pilots.  In one example,  a fire crew that was putting out a kitchen fire was suddenly told by the fire chief on the scene to stop and get out of the building.   The crew were puzzled but obeyed and as soon as the last man exited, the main floor of living room where they were standing collapsed turning the entire building to a blaze.   It had appeared to be a kitchen fire, but was a larger basement fire coming up through the kitchen and burning the floor underneath.    When questioned, the fire chief explained he wasn't sure what was happening, but that a normal kitchen fire should have gone out with the amount of water they had used.  In addition, the room was extremely hot for such a small fire with and unusually quiet. He reasoned something was wrong, there was an unknown danger and to exit.  Other studies have shown in the airline, medical, engineering and other professional industries that intuition from experience can play an important role.   Intuition is pattern recognition.  An experience person may not know exactly what a problem is but quickly determines conditions or events don't fit a normal pattern based on experience.   Where there is work at a site during initial surveys, construction, commission and startup, it is important to have experienced people involved that can see when something doesn't look right and raise an alarm.  The quote of having the best person at the site came from a retired successful project manager and considered by him to play a critical role in the success of any project.    When staffing a large global project,   having key people in the right locations are much like smoke detectors in a building.  Every risky location needs someone experienced enough to detect an issue early.   No smoke detector in a garage with flammable gas means there is a risk of finding a fire only after it is out of control.   The project is not better off by all the smoke detectors in the main living room.   When they all go off, yes, they are loud, but its too late.  An off-shore location, site location or other location key to the success of the project can quickly derail a project if problems are discovered too late in the project.  Once a project starts the FEL-3 or Front End stage, it pays to keep this same team through the end of execution.  Prior to this, projects may start and stop.  But stopping and starting a project between Front End Engineering and Execution frequently means the front end work will need to be re-done.  In addition to the general experience of people, there is value in maintaining experience through execution for automaton work.  

In May and June, 1940, the United Kingdom Royal Air force had suffered heavy losses over Flanders and France .  The RAF was outnumber by the Germans by 2:1 and only 620 operational fighter planes and well below a target of 1200.  However, Germany was still not able to launch any type of sea invasion without air superiority.  On July 16, Hitler launched operation Sea Lion for a potential air and sea invasion of Great Britain once the Luftwaffe gained superiority over the English channel.       A major project is somewhat like a war of information.   Huge projects and large global teams with emails and databases can generate so much information that the critical information is lost and the project drifts in the wrong direction.  Eisenhower was well aware of the issue and is well known for the grouping actions by importance and urgency to prioritize actions.   Urgent and important tasks are done immediately whereas urgent but not important tasks can be delegated. If it is important but not urgent, it just needs a date when it will be completed.  Anything else gets filtered out.  Every large project requires the planning and tools to deal with information and communication.  Large projects require plans so that the team is aligned with the process and doesn't get bogged down in non-critical information and overlooks information aligned with objectives.

Project success depends on managing timely information.  Sorting and reporting correct information  has been more critical as the amount of computer based project information continues to grow exponentially.   The Royal Air Force Sector Clock represents the historical  management tools used to achieve results successfully.  The clock played a key role in monitoring German Aircraft positions over Great Britain during the Second World War even though British Spitfires were outnumber 2:1 by the Luftwaffe.    Managing the limited defense plans and systems was critical and it was the Dowding system information that allowed decision makers to quickly react.  Aircraft positions were identified in 5 minute blocks associated with each color on the clock and reported to sector headquarters.  A plotting table at Bently Priory headquarters identified Friendly and Hostile aircraft with time-stamped,  red, yellow and blue moveable markers to give a real time view and lead time forecast for each unfolding raid along with defending aircraft.   A larger view of the entire country was assessed with key information passed on to Churchill's underground London headquarters.

Air Marshal Hugh Dowding was faced with managing the limit resources and in the end, thwarted the Germans by October 1940 that prompted Hitler to abandon the sea invasion.  The effort was heroic,  outnumbered Spitfire pilots with limited sleep would immediately fall asleep in the cockpit upon landing.    What gave Great Britain the edge was the world's first ground based interception network that gave decision makers they information they needed to quickly plan and react to German attacks.  Spotters and radar would provide information to sectors that would react with defenses.  The information was passed on to Bently Priory that would filter the information and display an overall map view updated every 5 minutes of the country based on all the information received.  Summaries were then communicated to Story Gate in London for any strategic decisions.  The tools at time included radar, clocks, charts, phone lines and spotters.   The Dowding system gave Britain the edge during this air war.  Without it, the limited resources would have been overpowered by Germany, the allies foothold to wind the war in Europe would have been lost.  The outcome would have been much different. The Dowding System highlights how processes and tools can make a huge difference in making the right decisions to achieve results.   

And as in NASA, the objectives are under schedule pressure.   Quality and procedure shortcuts can be taken.  If this is done early in a project, there is little consequence and the short cutting is reinforced.  It takes skilled and experienced project managers to know the difference between a short cut and improved efficiency.  On January 16, 2003, the Space Shuttle Columbia is launched. Only after 83 seconds into the flight, a suitcase-sized chick of foam from the main fuel tank and under the nose breaks off and strikes the leading left edge of the wing.   Aware of the foam breaking off, NASA engineers request the Department of Defense take an image of the wing but the request is not pursued by NASA management.   On February 1, the shuttle enters the atmosphere unknown to anyone that a 6-10 inch hole is in the leading edge of the wing.     During re-entry, heat destroys hydraulics  that cause the shuttle to lose control and disintegrate killing all on board. Followup investigations discovered that foam breaking off of the main tank had occurred in 3 other missions and in one case, put a dent in a booster rocket.  All flights were successful and in 2002, NASA determined the event of "foam shedding" was not a safety concern.  Some of the lessons from Challenger had not been fully learned.  In her book,   The Challenger Launch Decision: Risky Technology, Culture, and Deviance at NASA,  , Diane Vaughn describes how "Normalization of Deviance" can occur where deviations from normal or accepted practices become normal.  Once deviations occur with no consequence,  the lack of any serious incident further reinforces the behavior until a major accident occurs.   In both the NASA examples, there were plenty warning signs of problems that were ignored.   NASA was chastised for allowing the behavior to continue despite recommendations from technical team advisors.    Normalization tends to happen in large group settings where there is already an established "Group-think" that outweighs the concerns of a few.  New people who first see the issue are quickly "aligned" with the project and are only a small voice in the overall project.  It is brought up as a key risk to watch when moving to a small project to large project setting.  Projects fall into the same trap.  Many studies have been done to understand why large capital projects fail at an unacceptable rate.   Some of the quality processes that are frequently put in place or not questioned for applicability.  This is a frequent problem for automation projects where project quality assurance methods really applicable to non-software or mechanical projects are applied.  Managers think all is well until systems are integrated and problems start to occur.    The measurement of software development progress is very different than piping or mechanical design and construction.  Frustration can occur as people can't physically see automation software progress and can be surprised with results.  The answer to much of this has to do with testing by 3rd parties that verify software results along the way.   Software therefore takes on more of a "spiral model" of initial testing, integrated testing and site testing to find problems early.        And as in NASA, the objectives are under schedule pressure.   Quality and procedure shortcuts can be taken.  If this is done early in a project, there is little consequence and the short-cutting is reinforced.  It takes skilled and experienced project managers to know the difference between a short cut and improved efficiency.