Too many Change or Variation Orders

Asif, agility is indeed applicable to construction, there is the topic of lean construction, applying shop floor management to construction, PDCA continuous improvement and even general principles like disciplined agile (DA).

One major and starting point is the contract though, which should support collaboration, a joint responsibility to be successful, like with a consortium, or agreeing to iteratively review scope and price. Lump sum is preferred by buyers since it seems to put the risk on the seller, but in most cases leads to project failures. It is the old world that cannot survive.

Another option is to go for BIM, and manage changes using BIM, more efficiently and documented.

Here is a quote from McKinsey 2017:
Rewire the contractual framework. There is a need to move away from the hostile contracting environment that characterizes many construction projects to a system focused on collaboration and problem solving. To achieve this, tendering processes
can be based on best value and past performance rather than cost alone, and public processes streamlined. Establishing a “single source of truth” on projects for monitoring progress early, potentially supported by collaborative technology, helps to minimize misalignments and enable joint corrective action. The data already exist to fundamentally improve the accuracy of cost and schedule estimates. Where players continue to
use traditional contracts, they should introduce incentives that significantly improve performance and alignment not at a trade or package level, but at the project-outcome level. To move toward best practices, appropriate alternative contracting models such as integrated project delivery (IPD) help build long-term collaborative relationships. Relational contracts will need to become more prevalent than transactional contracts. Sufficient investments in up-front planning incorporating all parties’ input have been shown to raise productivity substantially.

Asif -

Some tactics I've seen used in the past are:

1. Rolling wave planning (where feasible) to avoid re-work due to requirements changing the scope or approach to a late stage deliverable.

2. Establishing a budget to cover a certain volume of change requests, especially when both buyer & seller acknowledge that requirements may change.

3. Having a "gatekeeper" to approve/reject change requests before the team commits to spending effort on analyzing them. On projects which are running low on time or money, the effort of analyzing change requests itself can cause a project to get into trouble so having this initial triage helps.

Kiron

Asif

As someone whose been in the construction industry for more than 15 years, I agree that change orders usually could drain the project.

Let’s first talk about what might be the root cause. I’ve seen this happen in the following scenarios:

1- The Design was completed without having clear vision of the scope and the desired outcome. In this case the client themselves weren’t sure what they want.

2- Lots of discrepancies in the design drawings and that could happen due to the complexity of design between Mechanical (HVaC and Plumbing), Electrical, Structural and Architectural. Those are discovered during construction only.

3- The design of the project was done without proper consultation and engagement with the customer and then after the fact, the customer starts changing things because they realize, this is not what they required for certain things.

4- The client keeps changing their mind about what they want.

The most common causes that I’ve encountered during my career are 1 and 2 and you can avoid such situations by:

1- In case there is no clear vision for the scope, try to do rolling wave planning (Agile Waterfall Hybrid Approach) and avoid going to fixed contracts. Instead, use CM at Risk contracts (In Canada it’s called CCDC 5B).

2- In case the design was complex, which for most projects these days it is due to complexity of the new systems, use a building information modelling program like Revit at the outset of the project. This 3D modelling will show most of the clashes between the different disciplines so designers can take that into consideration during design as opposed to being surprised by it during construction. This proved to reduce change orders by 80%.

Hope this helps.

RK

Define a project change management process, agreed and follow it. When people see that somebody will pay for changes then you will see how the amount of changes will be next to zero.

From related experience in large scale manufacturing, you might consider a dual-path approach:

1) Review of the change management process to determine whether it is effective and efficient. I have seen preliminary boards that screen changes prior to the formal review board for example, and we might collect multiple smaller changes until they can be incorporated together into planned "blockpoint" changes as examples of ways we have dealt with high change request volumes. That helps to cut down the high overhead that comes with the change process.

2) Collect data on the most common underlying causes of change. Are they missed requirements, customer requests, correcting errors, improvement ideas, etc. There may be some other process gap that is driving a high volume of change such as making certain design decisions before the data is available and a Just In Time approach is needed. Identifying the main causes helps preventing the need for changes in the first place.

Thanks a lot Rami, interesting and very much relevant input. I will surely further review and present these suggestions to our management.

Thanks Sergio, we have a well established change management process, but the point is to fix some kind of responsibility to those stakeholders who bring these too often.