Tips on using this forum..

(1) Explain your problem, don't simply post "This isn't working". What were you doing when you faced the problem? What have you tried to resolve - did you look for a solution using "Search" ? Has it happened just once or several times?

(2) It's also good to get feedback when a solution is found, return to the original post to explain how it was resolved so that more people can also use the results.

Unalloctaed Provision

6 replies [Last post]
Neil Brady
User offline. Last seen 11 years 20 weeks ago. Offline
Joined: 4 Jan 2007
Posts: 35
Whats the collective thoughts on Unallocated provision being applie once a Pertmaster analysis has been done.

For example, if your schedule risk analysis came out with a deterministic of 10% chance, and you deicided to use the P70 would you then add on unalloctaed provision ?

If you look at your P100 would you think that this is the maximum cost of a job ? If you choose to use it would there be no need for UAP ?

Lots of ways to put this across and I am very interested in how others apply this! as allows all advise is greatfully appeciated

cheers

N

Replies

Colin Cropley
User offline. Last seen 2 years 17 weeks ago. Offline
Joined: 1 Feb 2002
Posts: 57
Groups: None
Oliver,
I agree with your thoughts.
You can widen your closing statement to include the PM Maturity of organisations everywhere.
Fortunately, as good software becomes available, PM Maturity improves, as can be seenm from the adoption of the critical path method of planning.

Colin
Oliver Melling
User offline. Last seen 51 weeks 4 days ago. Offline
Joined: 24 Apr 2007
Posts: 595
Groups: The GrapeVine
Colin,

I have worked in several different companies and some have planning and estimating staff that never work together, whereas others do costed activity schedule estimates that can be used as the basis of the baseline programme.

I believe the best method is to use costed activity schedules to build a fully resourced baseline. This should then be fed into something like Pertmaster Risk Expert and impacted with uncertainty (3point) and risks (global risk).
As long as the activity settings are correct before impacting the baseline, a dynamic impacted plan can be produced that changes in both cost and schedule depending upon activity uncertainty, probabalistic branches etc.

The main problem i find is that estimates are always produced at the tender stage of a project and thus there is never enough time to run this analysis before submittal.

I think that such practice will eventually become mainstream, but it will only happen as the PM maturity of companies in the UK develops.
Colin Cropley
User offline. Last seen 2 years 17 weeks ago. Offline
Joined: 1 Feb 2002
Posts: 57
Groups: None
I have developed a methodology for cost risk analysis for a government authority using Pertmaster which I am currently extending to combined cost and schedule risk analysis using Pertmaster and our own risk management database, RiskIntegrator2, which maps risks to schedule activities in Pertmaster. This could also be done using Pertmaster Risk Expert’s Risk Register.

The basic approach is to deduct the [P50 forecast for cost (or schedule) excluding global risks] from the [P90 (say) forecast including global risks] to determine the Management Reserve, or Unallocated Provision (UAP) for cost (or schedule).

The logic for cost risk analysis is that the project estimate excluding global risks incorporates the contingency amounts for each line item and the P50 for the whole estimate (or perhaps technically more correctly the Pmean) after simulation, represents the Expected Value for completing the project incorporating the "known unknowns".

However, there are usually "global risks" identified in risk workshops that apply to the project, which are not reflected in the line item contingency allowances (three point estimates). These global risks with financial impacts and estimated probabilities of occurrence should be added below the cost estimate range analysis to represent the additional financial uncertainties to be taken into consideration in funding the project. Depending on the risk aversion or risk appetite of the organisation, the P80, P90 or even P95 project cost can represent the budget or bid value to be utilised.

A significant deficiency of the above approach is that it doesn’t take account of the financial impact of duration risk. The only way this can be realistically assessed is by integrating project costs and the project schedule by loading the project estimate into the project control or summary schedule and then performing a combined schedule and cost risk analysis. This is because a delay in a project cannot be costed unless we know in which activity or activities it occurs and also when in the activities, because of the profiled (rise and fall) nature of resource usage.

Similarly, when global risks with duration impacts that are not overlappping with duration ranges on existing activities in the range analysis are to be analysed, we cannot determine the cost impacts of those duration risks unless we apply them to a critical path schedule to determine whether there are project prolongation costs as well as individual activity delay costs.

By following the above described cost risk analysis approach for integrated cost and schedule risk analysis, we can derive realistic Management Reserves (or UAPs) for both Cost and Schedule.
Global risks are normally added to milestones of interest resulting in dual milestones: without and with global risks. Uncertainties both better and worse than the schedule and estimate assumptions can be incorporated by using probabilistic branching, with the branching percentages (which must sum to 100%) either symmetrical (50:50), or skewed to match the preferred or agreed likelihoods.

A challenge in having such an approach adopted is that the estimating and project planning processes are too often performed by different people in quite separate ways and often different stages of project development.

Colin Cropley
Oliver Melling
User offline. Last seen 51 weeks 4 days ago. Offline
Joined: 24 Apr 2007
Posts: 595
Groups: The GrapeVine
In commercially driven companies the very nature of business itslef is a political game.

Estimators/technical staff tell you how much the project will cost within an agreed tolerance +/-.

The technical base estimate is then used, and the tenders/bid departments will agree risk and contingencies monies to come up with what they deem a competitive price.

This may be done by putting in a lesser % for risk/contingency or by cutting the base estimate. (i.e using a P50 programme)

IMHO, when it all boils down, the price of a job is usually driven by the price of competitors bids!

So the question is, if the decision is made that no UAP monies are put in, does it really matter?

As a planner i would say it does, but it depends on the reasons that the PM/Commercial/Tenders have chosen not to.

Sometimes companies will ’buy’ a job to get their foot in the door. There may be strategic method to the madness.

So as a planner, i leave them to it.
Neil Brady
User offline. Last seen 11 years 20 weeks ago. Offline
Joined: 4 Jan 2007
Posts: 35
Chances of having a 100% risk impacted program ? No chance !

Chances of it being used as the baseline ? No chance if you actually want to win work.

Its a fine line.

I ask the question because within my company when doing a simple cost risk analysis within pertmaster (Or any other software for that matter) some people use the difference between the deterministic and the P50 (Or what ever is decided) as the basis for our UAP.

I am of the opinion that this is an inaccurate method of judging it.

Using Pertmaster to give you a probabilistic figure for your costs, gives you the % of confidence you have in delivering within a certain budget based on the risks of the KNOWN costs.

Again there is a bit of a debate, as it could be argued that anything unknown that adds significant scope is a change it doesn’t need to be included in the UAP.

That’s fine if its a client change, not so fine if its a balls up on your part. As you have said your probabilistic figure is fine if you can be 100% sure you have zero other risks on the project (very unlikely imo)

Now I realise that there is a political game going on behind the scene regarding bidding for work, competitiveness etc

But I would think that ignoring this if you choose to take the P50 you should have UAP on top of that to be safe

Am I correct in my assumptions ?

Advice greatly appreciated

Neil


Oliver Melling
User offline. Last seen 51 weeks 4 days ago. Offline
Joined: 24 Apr 2007
Posts: 595
Groups: The GrapeVine
It depends on how you estimate your UAP cost.

If it is a percentage of the total budgeted cost then it will go up if you choose to use a programme with a greater budget then your deterministic programme.

If you choose to use the P100 programme as the baseline and don’t have any UAP, then you need to be 100% sure you have captured and rated all the projects risks in you register.

Do you think it is possible to have a 100% risk impacted programme?
And do you think any PM would allow you to use it as a baseline!