October 27, 2007
As part of my PhD research, I participated in scheduling and controlling three construction projects in
Finland. The research plan was simple: educate the planners in location-based planning and controlling techniques, and then see how it works in practice with minimal intervention. All the projects managed to succeed in using location-based scheduling very well – location structures looked sensible, works were synchronized to optimize the schedule, and risk analysis was carried out to identify any weak spots in the schedule. However, all projects shared the same problem: there was never enough understanding of MEP. This lack of understanding manifested itself in several ways.
Planning and production phase problems:
Problem 1: Location Breakdown Structure
In many cases, the chosen Location Breakdown Structures did not match the requirements of the MEP workflow. For example, in a shopping mall project, there was a large shopping hall which had been divided to five areas according to how the structure would be erected. A location-based schedule was planned according to this location structure. When MEP contractors came on site, it soon became evident that work could not proceed in this way – the main ducts and pipes went straight through the shopping hall. When construction started, this fact resulted in the actual status compared to the planned, as shown in figure 1.
Fig 1: Planned vs. actual status of MEP ducts task in shopping hall
In figure 1, all locations seem to start almost at the same time. This is because each main duct is installed from one end of the hall to the other end before starting the next main duct. Therefore all locations have work in progress at the same time. Naturally, this forced all the succeeding trades to work in the same sequence, and the originally planned schedule became irrelevant.
Problem 2: Insufficient Quantities and Consumptions Information
There were very detailed quantities for all the other construction stages, but MEP always incorporated estimates for durations. Basically, all other trades were first optimized to have the same production rates, and to achieve the desired total duration, and then MEP lines were drawn in between with equal slopes.
The argument was to include this production rate in the subcontract agreement. After witnessing poor performance on site, I asked to check the contracts. The Flowline schedule was included in some of them, but there were no contractual milestones associated with production.
Problem 3: Level of Detail
The level of detail was usually insufficient, and many mistakes were made in the scheduling process. For example, the work of multiple subcontractors was lumped to the same task. The worst example was the task “MEP overhead”, which included the work of mechanical, electrical, plumbing and fire protection subcontractors and was planned with an estimated duration.
Problem 4: Wrong Links
Much of the logic between the MEP and construction tasks is soft by nature. It can often be changed or reversed, but there may be cost consequences in doing so. For example, in spaces without suspended ceilings, the roof can be painted before or after MEP has been installed. If painting is done before or at the same time as MEP, the ducts and cables need to be painted with the same color prior to installation. Another common example is MEP and slab-on-grade: It is more productive to install MEP from the top of each floor, but it is possible, with additional cost, to install from the top of the gravel.
Problem 5: Resource Constraints
When planning location-based schedules, many planners think that it is adequate to plan continuous Flowlines through locations. This works for many building trades, because they are specialized to work in just one Flowline task in the project. However, each MEP contractor has a large number of lines, and multi-skilled resources need to be balanced between all tasks going on at the same time.
A typical example in all three projects was that General Contractor requested more resources for a task, because the production rate was too low. Resources were added, but elsewhere in the project there was a delay. Instead of adding more resources to the site, the subcontractors moved their available resources from other tasks. The following week, the other task encountered a problem and resources were shifted back there. The problem could keep moving around unpredictably for many weeks before management understood what was happening.
Magnitude of Problems
The problems caused by poor scheduling caused slowdowns, start-up delays, interrupted work, and rework. In the simple five-million-Euro shopping mall project, a total of 111 production problems were identified. Of those, 76 problems, or 68 %, were directly associated with one or more MEP tasks. The other projects showed similar results. The interesting thing was that no one would admit that there was a problem. Actual costs were right on target, the final schedule was achieved without problems because of an excessively generous finishing buffer. For an outside observer, however, the inefficiency on site was appalling. If a solution could be found, MEP contractors would save huge amounts of money in labor costs, and buildings could be finished a lot earlier. I calculated that for the simple shopping hall at least one month could have been removed from the 9-month schedule if MEP had been planned and controlled more efficiently. This is 10 % in addition to the normal reduction of 10% that is achieved by implementing location-based planning!
Interviews of GCs and MEP contractors
Because this felt like the biggest unsolved construction scheduling problem to me, I interviewed both GCs and MEP contractors to see how they felt about the problem.
GC Point of View
Every GC reacted very defensively at first with some hostility from the MEP specialist. Then we started reviewing the data from their project. After a while, people started nodding: “Yes, I can see this happening”… “Actually this happens in every project”… “We do not have a clue”. I distilled the long list of comments into the following list of issues heard from General Contractors:
-
MEP tasks are long, dummy lines in master schedules
-
We do not have information about work quantities or required resources
-
MEP contractors actually tend to have a balanced amount of resources in projects. This results in production problems, because we plan unbalanced resource use for them (because of a lack of knowledge)
-
Location breakdown structures do not take into account the requirements of MEP
-
We do not have enough contractual methods of controlling MEP production
- Our own schedule performance is weak, so MEP contractors can always say: “work is progressing in parallel with the construction work” – because there are no resources in the schedule, it is impossible to determine the underlying reasons for any delays
-
MEP Contractor Point of View
MEP contractors agreed immediately that there is a huge problem with production control, and it is costing them lots of money. Electrical subcontractors in particular, suffer, because the bulk of their work is done in the final stage of the project. Production problems pile up and the electrical contractor is the poor guy who has to make up the delays of all the other contractors by adding huge amounts of resources in the final weeks of the project. These catch-up costs eat into their calculated margin. All MEP contractors say that productivity losses and uncertainty really hit their margins, because much of the work is done directly. The MEP contractor issue list is below:
- Resource needs are not balanced Resource needs cannot be predicted
-
It is difficult to control employees who work on a piece-rate basis (they maximize their own return)
-
The Location Breakdown Structure way of thinking is not widely understood
-
Production control happens only when the GC requires it
-
MEP contractor cannot trust the GCs’ schedules
- Tasks cannot start on time because of the preceding tasks, and cannot be finished
Trying to Find a Solution
It seems to me that both parties are feeling the pain over the same issue. The correct answer must be that GC and MEP contractors have to work together to solve these issues. The opportunity for the GC is a shorter duration of schedules, higher predictability, and smaller catch-up costs at the end of a project, and, in the longer run, lower bids from MEP contractors. The opportunities for MEP contractors are the direct cost savings associated with higher productivity and a lower amount of resources required. Also, the increased predictability of resource needs and starting times of activities have indirect positive cost effects.
I started to look for a group of companies (GCs and MEP contractors) working on the same project and willing to share information to develop a joint production control system. The idea was to have a common understanding and an agreement on the following issues:
In April 2007, I found the first such project in
Finland, and we started working together with the Helsinki University of Technology to develop a solution.
The results have been very promising, and another project has been started using the same methodology. I will report back on the projects as they progress.