(mis)Understanding Contingency

From the Risk & Reward Blog
by
I'm a risk enthusiast who likes to discuss techniques, tools and models—and use risk as a practical means to make better decisions in project environments. Join me on the ride!

About this Blog

RSS

Recent Posts

(mis)Understanding Contingency

Monty Hall for Projects: To Switch or Not to Switch?

10 Wishes for your Project Risk Management adventures in 2019!

The Curse of the Moving Mountain

A Template for Setting Up Your Qualitative Risk Analyses



Early this month I posted a poll here asking how you determine contingency and how much is generally considered.

In my venues, I usually map the risks and uncertainties as good as possible and run a simulation to understand variation regarding my goal, being it cost or schedule. I then determine a level of contingency for the project and add extra days or dollars to arrive at the percentile level.

Reading some references on risk analysis, people usually choose some references for safety. David Hulett, in his "Schedule Risk Analysis" (Gower, 2004) generally highlights P80. So I started wondering: how safe should your schedule be? How safe should your budget be?

My answer for that is that it depends. Depends on whether this project is another one in a set of 40 or 50, or if it is the only project of your firm, or the first project you are undertaking.

If you have a single project, you may be quite worried and stablish a high level of contingency, not to be caught off guard. I did some numerical experiments assuming a project would have an estimated P50 cost of 100 thousand dollars with a standard deviation of 20 thousand dollars, normally distributed.

If you want a safety level of 50%, you would always be prepared for spending the 100 thousand dollars for each project. However, as you add more projects to your portfolio, considering the results are statistically independent.

The question is how much should you add to each project in order to secure a safety level for the whole portfolio. If you have a single project, you need to add an extra US$ 32,900 to reach a 95% safety level. That's an increase in almost 33% of your budget! In the other ned, if you have 30 projects in your portfolio, adding US$ 6,000 in each project will get you to the same level of safety. 6% extra increase for a good night's sleep! If you want to see the detail, there is a graph and a table in the end of this entry with the data.

We should consider portfolio size in our decisions regarding contingency. That is why some references tell you to aim for the P50 or the average value in the simulation. If you have plenty of projects, the overall value will converge to the average in most of the cases.

Please let me know your opinion on this one. Do you run Monte Carlo to determine contingency or safety levels? Do you impose those values directly into each project? Do you have a line in your budget for the whole portfolio?

Looking forward to hear from you!

A graphical depiction of the example.

 

A table with the values calculated for the example

  Level of Safey required      
# of projects 50% 60% 70% 80% 90% 95%
1 100,0 105,1 110,5 116,8 125,6 132,9
2 100,0 103,6 107,5 111,9 118,1 123,3
3 100,0 102,9 106,0 109,7 114,8 119,0
4 100,0 102,5 105,2 108,4 112,8 116,5
5 100,0 102,2 104,7 107,6 111,5 114,7
6 100,0 102,0 104,3 106,9 110,5 113,4
7 100,0 101,9 104,0 106,4 109,7 112,4
8 100,0 101,8 103,7 106,0 109,0 111,6
9 100,0 101,7 103,5 105,6 108,5 111,0
10 100,0 101,6 103,3 105,3 108,1 110,4
11 100,0 101,5 103,2 105,1 107,7 109,9
12 100,0 101,5 103,0 104,9 107,4 109,4
13 100,0 101,4 102,9 104,7 107,1 109,1
14 100,0 101,4 102,8 104,5 106,9 108,8
15 100,0 101,3 102,7 104,4 106,6 108,5
16 100,0 101,3 102,6 104,2 106,4 108,2
17 100,0 101,2 102,5 104,1 106,2 108,0
18 100,0 101,2 102,5 104,0 106,0 107,8
19 100,0 101,2 102,4 103,9 105,9 107,6
20 100,0 101,1 102,3 103,8 105,7 107,4
21 100,0 101,1 102,3 103,7 105,6 107,2
22 100,0 101,1 102,2 103,6 105,5 107,0
23 100,0 101,1 102,2 103,5 105,4 106,9
24 100,0 101,0 102,1 103,4 105,2 106,7
25 100,0 101,0 102,1 103,4 105,2 106,6
26 100,0 101,0 102,0 103,3 105,0 106,5
27 100,0 101,0 102,0 103,2 105,0 106,4
28 100,0 100,9 102,0 103,2 104,9 106,3
29 100,0 100,9 101,9 103,1 104,8 106,1
30 100,0 100,9 101,9 103,1 104,7 106,0
31 100,0 100,9 101,9 103,0 104,6 105,9
32 100,0 100,9 101,9 103,0 104,6 105,8
33 100,0 100,9 101,8 102,9 104,5 105,7
34 100,0 100,9 101,8 102,9 104,4 105,6
35 100,0 100,9 101,8 102,8 104,3 105,6
36 100,0 100,9 101,8 102,8 104,3 105,5
37 100,0 100,8 101,7 102,8 104,2 105,4
38 100,0 100,8 101,7 102,7 104,2 105,4
39 100,0 100,8 101,7 102,7 104,1 105,3
40 100,0 100,8 101,7 102,7 104,1 105,2
Posted on: February 08, 2019 11:26 AM | Permalink

Comments (9)

Please login or join to subscribe to this item
The reserves are an economic amount that we add to the base line of project costs to obtain the budget, said in plain language is an amount that we reserve in case of any unforeseen event.
As in the case of contingency reserves, which is a single bag to cover contingencies, management reserves are an element that is above the project.

Interesting article on contingency determination for project. Thanks for sharing!!

Very interesting article. Thank you for sharing Guilherme!

I think it is important to clarify first what made up to the contingency in the cost estimates - does it includes funds to respond to risks identified in the risk register if risk occur and or allowance to cover the uncertainty in the estimate? In my view, both risk and uncertainty are two different things and has to be assessed through a probabilistic approach using 3 points estimates simulated within Monte Carlo. Some organization have corporate policy (that prescribed the required level of confidence in the estimates e.g. P50 or P80, etc.) in terms of budget control - for instance you may take the P80 on final estimates (to cover management reserve) but only allow P50 as budget authorize at project level.

In terms of multiple projects in the portfolio, I think it is not right to fix failing project from other project's fund unless such project is closed out and the fund is no longer needed...

Thank you so much, Rubén, Alok, Sripriya, Rajesh and Larry!

I am considering for this exercise that you have a baseline of cost considering an estimate with a given tolerance. On top of that, you add contingency in order to cover risks at an established percentile. That is my premise.

So, Larry, yes, I am including funds to respond to the risks identified in the register. In your last paragraph, you talk about failing projects. I would say those are costly projects. What I meant is that if you budget every project at P80, you'll probably have a big surplus of money in your portfolio, given that you have many projects. So it is one more thing to watch out for.

Thank you so much!

I think depending on how confident are we with our estimates, the ranges between the most-likely and the worst-case scenario in the 3 points estimates will influence the outcome of Monte Carlo simulation - the P90 result from the simulation wont exceed the estimated maximum values (for example if we assumed a base case of 100 units, most-likely 80% and worst-case 120%, I believe the P80 contingency outcome from Monte Carlo simulation wont exceed the 20% of the base value) for cost estimates uncertainty model (but of course risks cost model is different as we all know that a risk has only a 50/50 chance of occurring).

In terms of overall portfolio contingency sum, if for example I have a $10M project with 20% contingency, it doesnt make any difference compared to 10 projects of $1M with 20% contingency each&

I think I should emphasize the importance of modelling cost estimates uncertainty and risk impact cost separately. As mentioned above, risk impact cost unlike estimates uncertainty cost has only 50/50 probability of occurring (not all risk listed in the risks register will likely to eventuate) so let Monte Carlo give us a probabilistic risk contingency sum to be added to the estimates contingency, otherwise we'll end up with a big surplus of money.

Dear Larry, my point with the graph and the table was exactly to point out that, unless projects are 100% correlated, it is quite different having 10 projects of $ 1 M or 1 project of $10 M. If the projects are independent you will have a cancelation of surpluses and deficits, and the standard deviation will be much shorter than the single project experiment.

I agree with the separation of uncertainties and risks. There is a stablished uncertainty with estimation, for instance, and it does not refer to discrete risks. Those should be modeled on their own, and you should be able to see a Tornado of those risk evetns in order to prioritize efforts.

Please Login/Register to leave a comment.

ADVERTISEMENTS

"One never needs their humor as much as when they argue with a fool."

- Chinese Proverb

ADVERTISEMENT

Sponsors