Project trade-off analysis

Posted on 1st October, by JimYoung in Blog. No Comments

A “trade-off analysis” sounds scary or even criminal and customers might read it as “You mean I won’t get what I want?”  And yes, the customer may not get everything they want – it’s a matter of priorities.

A “paired comparisons” matrix is a useful tool for establishing project objective priorities.  To complete the matrix we have our client compare each of the four key project objectives against each other objective and in each instance record on the matrix the more important objective, then sum the number of times (score) each objective is recorded to determine objective priorities.  For example, we might find that our project is time-driven, in which case no work package should exceed a specific duration (say two weeks) to enable us to monitor with some accuracy schedule progress and take timely corrective action as need be.  Also, if our project is time-driven we would take particular care with risks that might impact project duration and if on-time completion is in jeopardy we might then trade-off the least important parameter in order to stay on schedule to meet our client’s top expectation of on-time delivery.


A trade-off involves a sacrifice that must be made to obtain a desired result.  During the planning or execution of a project there may be a need to alter one or more of the project parameters, but in doing so there will usually be an impact or trade-off elsewhere since key project parameters or objectives are interlinked.  For example, if we were to increase the project scope (ie, add work) we could expect a variation to at least the project schedule and/or budget to accommodate this change.

This means if we tinker with one parameter it has an effect on at least one other parameter, much like the following diagram attempts to illustrate.  You will notice that if the project is delayed, perhaps due to poor weather, the project cost is likely to increase.  This cost increase might occur because we usually need to pay wages and equipment hire charges even when work is temporarily halted due perhaps to poor weather.


There is a diminishing return and limit to how quickly we can complete a project by added more workers.  This diminishing return and limit is mostly due to 6 “C” factors including Coaching new arrivals, deterioration in work group Communications, Congestion at the work site, Co-ordination challenges, diminishing levels of Co-operation, and span of Control issues, the latter often resolved by introducing more supervisors, which of course simply exacerbates cost.  Brooks’ Law states “adding manpower to a late project makes it later” to which we might add “and also makes it more expensive and may even jeopardise the quality of the project deliverables.”  Final deliverables are what remain at project completion. They have their own life cycle, during which time they realise business case benefits –well that’s the intention.


Let’s consider the cost-time relationship in which scenario the project client is interested in the cost of accelerating the extra bedroom project from the normal 12 weeks at $61,000 to 11, 10, 9, 8 or even 7 weeks.  Seven weeks is the crash duration beyond which further acceleration will compromise the quality of the project deliverable.


The following table shows all relevant data associated with normal and accelerated task durations.  The last column shows the extra cost incurred for each week of acceleration.


A useful way to tackle such problems is to first prepare a network diagram, inset normal task durations, and thus determine the critical path that must be accelerated to achieve shorter project durations.  To reduce project duration from 12 weeks to 11 weeks the cheapest option is to accelerate critical Task D by one day, thus adding an extra $4,000 to the cost of the project. While we accelerate a project by speeding up critical path tasks, there soon comes a point when other paths also become critical and these too must be completed more quickly should we wish to further accelerate the project.

The following network diagrams show the most economical options to achieve the required durations.  I suspect in this example that the client would settle for one or two days acceleration, but beyond that point further acceleration becomes a very expensive proposition for this project.


The following series of networks shows how the project can be progressively shortened from twelve to seven weeks.  A task with an asterisk means it is at a minimum duration.


So $100,000 is an expensive extra bedroom, but for some time-driven projects, the extra cost may add considerable value.  For example, the amount of money that Oracle spent on modifications during the America’s Cup 2013, in excess of one billion dollars, was a case of speed regardless of cost.  Completely overhauling one of these boats in the middle of the competition is akin to buying the cup, which Ellison could afford to do it and New Zealand could not.  New Zealand had to pretty much stick with what they came with.  Skill in rebuilding is just as integral to this competition as skill in sailing.   Rapidly introduced technology won the race.  In future, a return to P Class yachts or a spending cap might be a good idea…


Comments are closed.


My latest thoughts on Project Management and life.


My big thanks to the coronavirus lockdown that provided me with an undisturbed opportunity to put together this 104 page booklet on Earned Value...

Project Estimating Checklist

Hofstadter’s Law: “It always takes longer than you expect, even when you take into account this law.”

Hofstadter was right....

Earned Value Management Misconceptions and Limitations

Although EVM offers excellent benefits, and is undoubtedly a very clever thing, it’s not perfect. In this final chapter I’ve discussed in detail a...