Saturday, January 23, 2010

Work-Arounds in Haiti

A good friend of mine, Mike Wnek (pictured above), was down in Haiti shortly after the devastating earthquake hit there on January 12, 2010. His story was well-documented in the news (click here, here, and here), as he and his contingent were one of the first groups to deliver desperately needed food and water to the ravaged city of Port-Au-Prince.

While tons and tons of supplies and resources were being bogged down at the airport in Port-Au-Prince or being re-routed to neighboring nations, Mr. Wnek was riding shotgun on a rented truck as his group made their way from the Dominican Republic to the Haitian capital with a truckload of survival supplies. Only their courage and resourcefulness made it possible for them to successfully complete their voyages across Hispaniola. Their actions were truly remarkable.

Once I had time to reflect on what my friend and his group had done, I began to assess their actions in an objective way (this is easier said than done when people are performing heroic deeds to save lives). Once I had assessed their actions from a business perspective, I realized that they had made great use of one of the all-time most frequently used tactics--the work-around.


Pretty much anybody who has ever worked in an organization has had to resort to work-arounds to get things done. Work-arounds are informal, alternate methods for completing work that frequently arise in response to ineffective formal work processes. Basically, whenever doing work the normal way is just not practical, we resort to the path of least resistance.

Folks in the the Lean world understand this concept all too well, as the presence of work-arounds is one of the best indicators of poor process flow. Sometimes a process is unreliable, thus requiring an expert to babysit the work. Sometimes there is too much buffer between processes, thus requiring an expediter to push the work to the front of the line. There is no limit to how many work-arounds we might find in a workplace.

Work-Arounds on Projects

On projects, every bit of the work might be completed through work-arounds. This is quite common in organizations that are not organized for project work. These types of organizations typically feature a traditional vertical hierarchy, and are usually geared for only ongoing operations, not temporary projects.

The unfortunate soul who is assigned to lead a project in this type of organization usually has to navigate a byzantine web of work-arounds cutting across several vertical silos to get anything accomplished. This sounds a lot like what my friend, Mike Wnek, had to do in Haiti.

Quick Fix in Haiti

When it became apparent that the most obvious needs of the earthquake survivors, food and water, were not being met by the official agencies in charge of the relief effort, Mr. Wnek and his group immediately constructed a makeshift work-around to get aid flowing. They went to small grocery shops in the Dominican Republic and bought their entire stocks of food and water, and then proceeded overland to deliver the goods.

Forget the airport. Forget the air drops. Forget bureaucracy. Forget red-tape. Just get food and water to the survivors! This was a classic example of the use of work-arounds, albeit in a highly remarkable context.

Learning from Haiti

Why did my friend and his group have to resort to these desperate work-arounds? Why didn't the aid start flowing into Port-Au-Prince sooner from big relief agencies? Why did the focus early on seem to be on riot control instead of the delivery of survival resources? Why were evacuation efforts diverted? Why was the airport bogged down? Lots of questions.

If we can say that disaster relief efforts are a form of project, then we can probably deduce that many of the same factors that we see on everyday projects are in play on relief efforts as well. Maybe we have vertical silos that aren't communicating with each other. Maybe decision-making is not being made by the people who understand the situation on the ground. Maybe not enough contingency planning had been performed (there is no time to plan a relief effort after the fact). Many factors could have contributed to the delays in Haiti, but there's no way to completely understand the situation from afar. Maybe an expert in large-scale project management will someday publish a case study on the response to the Haitian catastrophe.

Moving Forward

Understanding the lessons learned from Haiti is critical. Without it, we are doomed to repeat the failures of Haiti, Katrina, and elsewhere. But we can not blame individuals! The people working on disaster response projects deserve our respect and appreciation. Just as on everyday projects, the system is usually the culprit, not the individuals. We need to have better project delivery systems in place to support the efforts of individuals.

If we don't improve our systems, we better pray that we never find ourselves in the middle of a natural disaster. After all, there are only so many people like Mike Wnek who have the ingenuity and courage to establish a lifesaving work-around supply chain in a disaster zone. Do you want to rely on heroics?

Monday, January 04, 2010

3 Lean Tools for Improving Construction Reliability

The Opportunity

“Let me get this straight, you think we should reduce our inventory of cabinets? That’s our safety buffer! What are we supposed to do when the hot glue machine breaks down again and we can’t build cabinets, huh? We gotta have a buffer.”

That was the reaction I got from a construction manager upon hearing my suggestion that he winnow down his inventory of finished cabinets (this was on a construction project where the cabinets were being prefabricated nearby). His response was valid, but so was my suggestion.

From my perspective, I saw several problems with the excess inventory—problems that are usually associated with this form of waste: finished cabinets were getting damaged as they sat around, they were often in the way of installers working in the building, and they were even creating a trip hazard for any passersby. It was not pretty.

From the perspective of the construction manager, he saw any reduction in cabinet inventory as a risk to the project schedule. His point was that if he eliminated his safety buffer, the unreliability of the cabinet-building process could cause cabinet production to halt, and potentially cause construction delays. Not only did the cabinet shop folks have problems with the hot glue machine, but they were also dealing with a whole host of other issues that created variation in process results: untrained cabinet builders, a messy workshop, conflicting production schedules, etc. Again, it was not pretty.

The Lean Approach

So, what to do? What would you, as a lean thinker, do to create more reliability in the cabinet-building process? I’m certainly not an expert in the lean methodology, but I have seen some practical lean tools applied to construction processes that have yielded improved levels of process stability. Here are 3 of my favorite such tools:

1. Tool # 1—5S Visual Workplace

We all know about 5S by now, I’m sure (check out JC’s excellent blog post). It’s a fantastic tool that can really help us create visual control of the workplace, so that we can spot abnormalities quickly. This would help reduce variation in the cabinet shop by reducing time wasted looking for tools and materials, as well as by eliminating the chance for a lost-time work accident by removing safety hazards from the workspace. This would no doubt help improve the stability of any construction process.

Construction Industry Particulars...

However, applying 5S to a construction site is a bit different than in a traditional workplace. The main difference is that construction projects are temporary in nature, while ongoing operations tend to be a little more permanent (although a wise strategy would be to design the workplace to be flexible enough to change with the times). This means that a 5S process designed for construction sites would need to be able to be implemented during a short ramp-up period, flexible enough to accommodate multiple stages of construction, and exceptionally easy to understand for those random people that tend to visit construction sites (inspectors, salespeople, etc.).

What other barriers exist to using 5S in construction?

2. Tool #2—Preventive Maintenance

Just like factory workers depend on conveyor belts, press machines, and welding robots, construction workers depend on their ladders, generators, and hand-tools. Unlike lean factory workers, most construction workers do not perform much preventive maintenance for their equipment. Often, equipment is just loaded up at the end of a long work day and tossed in the work truck. Worse yet, construction equipment is often exposed to Mother Nature in ways that most industrial equipment is not. This leads to a ton of equipment failures that slow down our construction processes, much like the hot glue machine did for our cabinet shop. A great approach for mitigating this source of variation is to implement basic Preventive Maintenance (PM) procedures.

Construction Industry Particulars...

PM is not a difficult concept to explain, but a frustratingly difficult tool to implement in the construction industry. This probably has more to do with bad habits than anything, so the big challenge is creating a work culture than encourages good habits. Instead of making it a habit to knock off work at the last second before the sun sets (which leaves little time for Preventive Maintenance), we should build-in time to inspect and repair our equipment. Instead of making it a habit to drag equipment on the ground, we should make sure that our people have better ways of transporting heavy items. Bad habits are hard to break, but the benefits of having reliable equipment far outweigh the costs of making a cultural change.

Do you think it would be too hard to get buy-in for PM based on your experience?

3. Tool #3—Job Instruction

Job Instruction (JI) is another great tool for improving process reliability. With construction processes, a lot of variation comes in the form of different techniques being used by different installers. Often, you can observe two installers building cabinets in four different ways. This is not good for consistency.

JI helps mitigate this source of variation by providing us with an effective approach to teaching standardized processes. Once we have established a set of best practices for cabinet building, we can incorporate them into a job sequence that can be taught using the Job Instruction method. This is a great tool for training new installers and cross-training veteran installers. By properly training our people, we can reduce variation between installers and greatly reduce the chance of human-related errors occurring.

Construction Industry Particulars...

The difficulty of implementing JI in the construction industry is that quite often the work is being done by sub-contractors who are not always amenable to being trained. Their business relies on having a reputation for knowing how to do good work, so nobody wants to submit to training, as that is an indication that they are still learning how to do good work. This is a huge cultural, systemic issue in the construction industry.

One way to overcome this barrier is to take the Toyota Way approach of investing in long-term suppliers. Choose sub-contractors who are open to long-term learning and partnership. Provide them with training on how to perform JI, and let them become their own trainers. Work together to develop the standard job sequences that are being taught. Include JI as part of the statement of work for the contract. Think long-term.

Do you foresee construction managers allocating time for JI?


The three tools listed above are just a few of the many available in a lean construction manager’s toolbox. What other tools might you recommend for our cabinet shop? Do you think the tools I listed are appropriate given the situation I’ve described in the cabinet shop?

Of course, the particular tool chosen is not what’s important; it’s the thought process that counts. The thought process should be to look for barriers to creating better process reliability, and then pick the right tool to address those barriers.

Eventually, as we stabilize individual processes, we can begin to reduce our buffers and improve the flow between processes. This leads to problems being surfaced more easily, and to waste being systematically eliminated from our processes.

In the case of the cabinet-building process, if we were to implement 5S, Preventive Maintenance, Job Instruction, and other appropriate tools, we should see an improvement in process reliability. This would allow us to lower our inventory of finished cabinets at the job site, which would in turn eliminate the wastes associated with excessive inventory. This would lead to what we all lean thinkers want—better results.

Do you agree with my hypothesis? Do you have an alternative approach?