Many of the struggles we are currently experiencing when attempting to implement Lean in the construction environment are the direct result of applying Lean tools out of their proper context. Understanding Lean as an operating system will help you to avert this all-too-common pitfall.
As discussed in the Introduction, the annual Industry Week Census (released in 2007) reported that 77% of manufacturing plants surveyed were utilizing Lean as an improvement method. Of these, 2% reported that they had achieved World Class Status, 24% reported significant progress, and 74% indicated that they had attained some or no progress. These are not exactly stellar results. And there is no data to suggest that Lean implementations in the construction industry are enjoying any better success. Given the multiple players and competing interests that come together to produce our product, it isnât difficult to extrapolate that the success rates for Lean in the construction industry are even lower.
But even this is difficult to ascertain. In his incredibly comprehensive thesis, âMeasuring Lean Construction: A Performance Measurement Model Supporting the Implementation of Lean Practices in the Norwegian Construction Industry,â published by Norwegian University of Science and Technology in June 2015, David Herranz Limon provides an extensive review of current Lean theory (Last Planner, Pull Scheduling, Concurrent Engineering, and Virtual Design Construction) and measurement methods (Balanced Scorecard, European Foundation for Quality Management Excellence Model, Key Performance Indicators, and Lean Six Sigma) and concludes that given the âlack of measurement cultureâ as exists in the construction industry, and with so many variables at play, it is difficult to precisely state what improvement gains are specifically derived by employing Lean methods.
Those of us who have witnessed labor rate productivity improvements, cost reductions, quality improvements, and schedule enhancementsâall of which were the direct result of targeted waste identification and elimination effortsâcan easily point to quantifiable gains as a result of implementing Lean. Recently, a CEO we worked with said, âWe are having the best top and bottom line year in our history. When we engaged with you we had come off of a tough year earning only $450,000 EBITA on $128M revenue. This year we are on track to earn $15.5 million (!!!) on $270M revenue.â Clearly, not all of this was due to their Lean implementation, but it certainly was a contributing factor. Yet the niggling sense that Lean is not making big âbang for the buckâ inroads remains. Is this just a matter of statistics and of finding the right data points to âproveâ Leanâs validity and success rates? This is an interesting question and one most vexing for the construction industry. With so many variables at play, this may prove to be a fruitless quest, though this hasnât stopped the engineers among us from trying.
Part of this quantitative quandary is due to the fact that, in most cases, Lean is implemented out of context. Instead of being applied as an operating system company wide, meant to eliminate waste in the office and the fieldâfrom Request for Proposal (RFP) to Project Delivery or ServiceâLean is often applied in piecemeal fashion: Last Planner on a project here, Value Stream Map (VSM) on a process there. As such, it is harder to gain a sense of what Lean is doing for a company as a whole.
But I think there is another issue at playâand it has more to do with the human element than finding the right quantifiable measure. As Neil Postman states in his book Technopoly, we have become far too reliant on data and technology to guide our decision making and assessment of the effectiveness of various improvement methodologies. His contention is that âwe live in a self-justifying, self-perpetuating system wherein technology of every kind is cheerfully granted sovereignty over social institutions and national life.â Though I am a huge proponent of basing decisions on objective rather than subjective data, I believe, at times, practitioners of Lean and its first cousin, Six Sigma, are overcompensating for a lack of measurement in our industry by instituting overly sophisticated statistical analysis in order to justify Lean methodologies. As a result, we have inadvertently contributed to what Postman describes, as a ââŚgrand reductionism in which human life must find its meaning in machinery, measurement and technique.â As he asserts, in so doing, technology supplants culture, and in our quest to define what is effective, we inadvertently reduce people to machine-like entities while elevating our view of machines (in particular, computers) as some sort of âidealâ that people should aim for, i.e., able to make reasoned, rational decisions at all times, based on data that we define as relevant. Allow me to point out just how wrong-headed this approach can be when applied to construction. This example is extracted from one of my assessment reports:
As much as we love measurement in Lean, it is possible to have too much of a good thing. And at XXXXX, you have truckloads too much. You are drowning in metricsâand many feel this is creating more waste, rather than eliminating it. Numerous people are wondering loudly about how much it is costing the company to generate reams of data and information that virtually no one usesâand worseâthat most see as counterproductive. Let me give you one example.
Currently, the company is tracking overtime usage and publicly ranking field people in terms of overtime usageâthe assumption being that overtime is a wasteful expenditure and is the result of poor planning. Iâm sure this is the case at times. But this can in itself be an erroneous assumption. Overtime can also be caused by:
An owner that makes numerous changes, yet due to their proforma, needs to hold to the same end date, thus dramatically compressing schedules. If they are willing to pay for overtime and view it as value added (and are, in fact, demanding it to stay on schedule), why would this not be factored into the rankings? (Currently, it is not.)
Market conditions, i.e., when other trades that are piecework driven provide incentives for workers to stay on the job for additional hoursâthus putting these trades ahead of schedule. If the superintendent or foreman allows their job to âget buriedâ by these other trades, this company will incur increased back charges for damaging their work, or slowdowns while attempting to do workarounds. And the impacts of these slowdowns will increase the further the work falls behind the other trades. Overtime, in such instances, may better serve the system by preventing waste.
These rankings also seem to ignore the role that internal design, engineering, and estimating play in our system. After all, the field is merely the repository for all of the other broken process pieces that came before them. Thatâs not to say that the field doesnât have its own role to play in terms of waste, but I donât understand why field people would be singled out, when clearly this is a systems issue.
Rankings such as these often drive a stake into the heart of teamwork. Why would any superintendent or foreman send any of their guys to help out other projects if it meant, by doing so, it could result in a higher overtime ranking for the person they helped out and a lower one for themselves? Measures like this inadvertently add waste, rather than eliminate it, by discouraging collaboration and teamwork. I know that your qualitative analyst believes he accounts for such factors under the umbrella of âexceptions,â but in reality, he does not. If you are a foreman or superintendent whose job is currently beating the projected budget, yet find yourself ranked at the bottom of overtime usage, this data is demoralizing and pointlessânot fruitful or instructive. Further, it will lead them to resist further usage of metrics, or encourage them to provide false data, in an attempt to improve their own metrics. All of this is counterproductive to a team environment.
Lastly, what is the end game of ranking field people? If overtime were the result of poor planning, I donât understand how shaming people is going to help them to improve. As stated above, I think this will have the opposite effect. Rather than seeking ways to improve, people will resort to not-so-productive ways to avoid shame. Wouldnât it be more prudent to use this data to mobilize the management team to bring company resources to bear and come up with a plan to help them improve?
Is there any wonder why people sometimes roll their eyes when we utter the word Lean?
Lean isnât about perfectionism, though sometimes people do apply the notion of continuous improvement as a way of feeding their obsessive-compulsive tendencies, and thus end up driving everyone else crazy in the process. Perfectionism is exhausting and demoralizing. Continuous improvement is about optimismâthat we can exert control over the things we can control and make our world a little better every day. At its core, Lean is uplifting and motivating. And when we come to realize that it is through seemingly small acts, such as saying thank you, recognizing the actions of others, taking the time to explain something, or truly listening to what a person is struggling with, that we are making a contribution toward making all of our work lives just a little bit betterâevery single day.
One of the joys of being in this industry is that it is dominated not by machines, but by people. Itâs people that do the work: weird, quirky, diverse, wildly intelligent, dumb as a bag of rocks, zany, funny, sometimes downright scary people. Unlike a mechanized assembly line, we canât fully âerror proofâ our projects. So, that means all of us are stuck dealing with our messy, sometimes irrational, flesh-and-blood compadres. Some Lean practitioners try to get around this by doing the next best thing: standardizing as many repeatable practices as they canâattempting to âidiot-proofâ our job sites in the same way that McDonaldâs idiot-proofs the keyboards at their cash registers.
Donât get me wrong; I actually practice a branch of psychology (cognitive-behavioral) that focuses on isolating independent from dependent variables, demands statistical analysis, and subjects findings to peer review and replication to cull fads from practices that are empirically sound. So, I am a big fan of the scientific method. A...