Project as a Complex System – Feedback loops

Posted by Girish Deshpande in

 Service Excellence & Methodologies

Dec 15, 2011 | Total views: 171

1

Link

Girish Deshpande
Technical Director - Industrial Systems and Smart Energy Group
More articles by the author

In my last blog, I said that a typical project is nothing but a complex dynamic system. At least, it shows all the characteristics of the complex system the way any other complex system typically would do. If the project is a complex system, then we need to apply systems thinking to understand what is going on with this system and come up with the solution to the problems we witness in such systems.
A couple of things that we need to understand when it comes to systems thinking are:

1. Preference for thinking in circles over straight line (cause- effect kind of thinking)
2. The connections between the parts form the vital feedback loops.

The feedback is the output of the system re-entering as an input to influence the next step, positively or negatively.

The feedback can be of two types:
Reinforcing feedback: The changes in the system re-enter and amplify the very change, leading to more change in the same direction. This can result in runaway effect, either a growth or decay, depending on the type of change. In the project context, we see such runaway conditions developing when the team compromises on quality to meet the deadline. This results in customer escalation and rework which puts pressure on the next deliverable. The team may decide to compromise again in order to meet the deadline because the customer will be upset if we miss the deadline again. They have to choose between the devil and the deep sea. Both are uncomfortable choices for the project manager. The project is going out of control. On the other hand, the positive example of such feedback is the early project success that boosts the team morale and confidence and builds upon itself for the team to deliver the project on high note. The reinforcing loop is the same, but the outcome is different.

Balancing feedback: The change in the system is fed back to oppose the same change and so, dampens the effect. It reduces the cause that is creating it. Balancing feedback keeps the system stable and resists an attempt to change it. All complex systems have a goal, typically it is its balanced state, and the balancing feedback reduces the gap between this desired and actual state. This is how it takes the system towards its goal. In the project context, this is what we call as tracking and monitoring mechanisms. The examples of this can be project status reviews, rapid prototyping, product testing, and team appraisals. You are trying to detect the issues or new risks early and take corrective actions to avoid getting into runaway conditions described above.

But there is another worrisome example of the balancing loop. The balancing loop can also pose “Limits of Growth” in a given system. The success in one part causes counter balancing effects that restrict the further growth of the system. The phenomenal growth of the company causes other complexities and problems that restrict further growth. No tree grows to the sky. Typically you apply more pressure on the reinforcing loop to break out of this cycle; like doing more of the same or doing more of what has worked in the past. But it rarely works. Incidentally, such pressure on reinforcing loop only strengthens the balancing loop that is limiting the growth. The solution is to identify levers in the balancing loop and weaken them to improve the performance. In the project context, the example can be, to add more people when the schedule is slipping thinking that more people will solve the problem. More people create more co-ordination issues that caused the schedule slippage in the first place. If you do not address the real cause of the schedule slippage which might be, for example, integration problems and resulting quality / rework issues, then adding more people will only make the problem worse.

When it comes to feedback loops you also take into account the “law of unintended effects”. When we strengthen one aspect of the system, it can have unintended effect on other parts and eventually on the entire system. That is why systems thinking focuses on the entire system and not on individual parts alone. It values holism to understand the effects of any action that we undertake on the system. This problem is commonly seen in incentive systems adopted by the organizations. The incentive programs encourage certain kind of behaviors from the people but that can have unintended impact on the other parts and hence, on the entire system. For example, if the project is judged on its profitability and the PM rewarded for this performance, then it can encourage the PM to game the system by not staffing some team members on the project or not staffing them 100%. If the business development team is incentivized based only on the revenues, then the quality of those revenues will go down. There won’t be any incentive to align the new business with the organization’s strategic goals or profitability of the projects.

It makes sense to understand the effect of any action on the entire system by applying circuitous systems thinking. Thinking in circles pays.