James,
thats great if you can find enough Clients who can afford the target cost and are willing to come back to you time after time when you fail to meet it.
The rest of us just find that the Client refuses to pay above the target cost and moves on to a competitor for the next job and asks for some money back to cover his expenses/costs involved in your failure to meet what was promised.

Its nice to know that some pockets of the industry are still shielded from reality!

James,
well how do you stay in business?

Perhaps your management have higher considerations besides budget control and manpower resourcing to make a profit (thats if you need to make one)

Didn’t keep an old timesheet as an example unfortunately for bookings to drawings.

Your point about Management control is the key - if you don’t manage manhours effectively how do you stay in business?

The budget for each discipline & work package is set from the sold hours. The PM opens only those WP’s which are being worked on and the guys have to choose an open WP code when booking. We get weekly reports on Monday am at the WP level. The Lead Engineer has total responsibility thro the PM.

Peter,

I agree with the principle behind your statement. Moreover, I’d be extremely interested to see an example programme where the DO engineers were booking their time against each individual drawing. Please, please send me one!

Four years ago, the company introduced Project Central. This has the potential to allow Engineers to book their time at whatever level of detail was required, directly against task descriptions that reside in the project programme. The system then automatically incorporates all their timesheets - thus saving absolutely enormous amounts of time. In fact, what it does is disperse the "timesheet data input" across the whole team. The problem we encounter, though, is that the individual engineer doesn’t record his time particularly accurately. It’s difficult enough to get them to book accurately at the sub-assembly level, never mind the individual drawing level. Unless we can get a complete cultural change, all you’d be doing is dead-horse-flogging. Unfortunately the management culture also seems to pay little more than lip-service to planning & control.

Peter: your experiences please.

James.

Gareth,

EV is a perfectly good method for use in the Design environment, on the assumption you use the appropriate measurement unit. We use man-hours, owing to the fact that it is the most straight-forward, is equally as accurate (or inaccurate as the case may be) and minimises the administrative burden of measurement. We do not attempt to derive the number of drawings and a weighted man-hour value against each one due to the fact that it would be so onerous and not any more accurate. However, probably the most significant elements of project progress calculation are the levels of detail at which you need to report, or wish to dissect the EV, and the accuracy at which you want to monitor.

Many, many clients, who are predominantly from the “construction” arena, still have this mental block about using man-hours as a form of measurement, and they attempt to impose a “physical quantity” measurement unit – usually: “quantity of drawings”. But is just doesn’t work. This is mainly because, at the outset of any design process, you just have no real idea as to how many drawings are going to be produced. This is especially true in our environment because most of what we do is one-off custom engineering. The construction industry may be different because they have a lot of historical data upon which they can call. Moreover, there is probably a standard correlation between the size of the building and the total quantity of drawings produced.

Our current, basic process is: Produce 3D model – Check Model Integrity – Produce Initial Drawings – Check Initial Drawings – Update Model – Update Drawing – Final Check of Drawing. In parallel with this process, there are ongoing structural and engineering calculations, Design Reviews and Manufacturing Reviews, all of which induce alterations to the models and drawings. This is generally programmed at the sub-assembly level, using man-hours only to deliver a suite of drawings (approximately analogous to a single room of a small, multi-story office block. This could be considered excessive, but I am performing an experiment to determine where the real time is spent, thus improve our estimating capabilities.). However, we only calculate the EV at the Assembly level (analogous to a complete floor of the office block).

The next logical level of detail would be the individual drawings. However, at the next level down, this would probably increase the programme size by anywhere up to a factor of 100 – which then requires an army of administrators to manage. This is due to the fact that, if you are using the quantity of drawings to measure the progress, you would have to know where, in the basic process, each one of those drawings was and have calculated an approximate value for each step of each drawing. Bear in mind that some drawings will have ten-times as much work as others.

Ultimately, you want to keep it as simple as possible, consistent with your needs. In the end, everything is merely an estimate, so you won’t want to get bogged-down in minuscule detail that serves no purpose. Moreover, when you examine an EV curve (SPI or CPI) you are looking at the TREND of the curve, not the accuracy or precision of the absolute value.

As an aside; point 4 of Peter’s posting is an interesting one. Thinking is a very important part of the design process – albeit it an intangible one. We can all scribble lines on a paper and achieve an output – but at what level of quality? Spending 50% of the budget just thinking about something – before you put pen-to-paper – could save you thousands later on. This is one of the flaws of an EV system in that it rewards physical output regardless of the quality. This is partially why using man-hours is very useful – because who gives-a-stuff how much of the time you spend thinking about it, relative to pressing buttons - as long as it costs you exactly the same. However, if you have to build another one, your costs should be substantially less because most of the “thinking” costs won’t need to be incurred.

Please feel free to raise issues that require further clarification.

James.

Gareth,

I’d imagine that Architectural Design utilises the same principles as any other “Design” process, with the initial output being a set of drawings from which the physical entity can be manufactured or built. If you’re happy with this interpretation, then I’m more than happy to waffle-on with regard to my six years’ experience of using EV in the Design Planning environment.

Let me know.

James.