Simple. The model that will have more credibility is the one the estimator will bet their liability insurance cover on and present as the firms "official" estimate to the client. And also if they will not stand by it with a guarantee of accuracy, what use is the BIM advantage?
Estimators who base their bids (or should that be bets? )on the basis of the Bim model are wagering that all elements in the model have been categorised correctly, that the information assigned to each element when entered by the technician is complete, correct, or accurate for pricing and that every thing you are looking at in the model has bee drawn and measured in 3D and not "worked around" in a series of 2D views.
How would I expect the owner of an estimating firm to verify what the schedules show? I'll bet that the most cautious run a parallel estimate.
Also how many estimators do You know who are not risk adverse (read brave)?
Kevin
I am in the middle on this one. As I have said in other forums, I have never been asked by a client or contractor to furnish a BIM whether we offer one or not, and I would not expect to be able to bypass drawings in pdf and hardcopy if I were. If (when?) I do have that opportunity, I would look for some middle ground. I am a BIM believer, but I would not expect a construction company to base their hard-bid on an un-audited model. Were I in the GC's shoes, I would want some assurance of what was and was not modeled, what conventions were used, and some recourse if the model did not support those assurances. Who would do that audit, and what form it would take I am not sure. What kind of recourse? Not sure about that either.
I see the point of this article, and I agree that pure parallel processes are redundant. But -
A contractor who is new to BIM, and takes the leap of faith will need to expend resources to
- obtain the needed software
- train several staff to estimate and/or schedule from a BIM
- agree with the designers on the appropriate level of completeness for the model at given stages of the process
- agree with the designers on formats and conventions to be used
- have the model audited (at each stage of model development)
- perform "pick-up" operations on the model where the design modeling does not meet specs (or does meet specs, but does not meet needs discovered along the way).
Don't these have the (admittedly and uncomfortably unpredictable) potential to exceed the resources required to run parallel processes?
I recall parallel processes, which often ran for a full fiscal year or longer, being de rigeur back in the '60's and '70's when businesses moved from manual to automated processes. The change to BIM is not quite on that scale, but it is major. The thought that the new process will fail altogether, or in some crucial aspects, or for some period of time, lurks in the back of an adopter's mind. They would be irresponsible to ignore that thought.
It is only reasonable and prudent for a GC new to BIM estimating and scheduling to
- run checks on significant quantities and time
- backup the BIM-based estimating and scheduling at major milestones so that a glitch does not bring the whole process to a halt for a few days.
I doubt that there is any way for overhead costs for the first few BIM projects to be lower than they were for the last project done conventionally. So the ROI on the first one is almost bound to disappoint. Sorry, but that is my sincere and honest opinion.
Thereafter, I firmly believe that confidence, proficiency and efficiency in the use of BIMs will grow quickly. But, as we have seen internally, there are sticky places where the software is (after 10 years of upgrades) not quite able to do what we want it to. I am not talking about exotic parametricity or eye-popping graphics. I am talking about consistency of attributes from one door or column to another, or the need for "workarounds" that are necessary to produce clear drawings, but foul up the model and therefore impact BIM-based price and schedule. Those kinds of anomalies in a model have to be dealt with off-line (basically, manually). I doubt that they will ever be eliminated entirely, but their numbers are declining and that will continue.
Clear-eyed estimators and mangers with sound experience-based judgement will be more productive, but no less in demand than today.
Two caveats:
- I do no doubt that the end results are better when all parties are fully committed to BIM. An all-in bet is right sometimes. But there is no bluffing in construction bids.
- Olli Seppanen has literally infinitely more experience applying BIM to construction processes than I. So it is likely that he is right and I am wrong.
My experience is in application of BIM to architectural and structural documentation processes. The first time out 13 years ago, I was not efficient. I was learning. I went to classes. I made numerous trips to the user manuals. I made mistakes. I corrected those I could correct in a reasonable time. I found limitations in the software I was using. I had to use workarounds, I needed to devise (with help) new graphic standards and documentation techniques. The end product, while acceptable, was a departure from the standards used for 2D CAD-generated drawings. Overhead costs were ridiculous on that project. The 20th time out, on a larger and more complex project, the documents (and the model) were not perfect, but quite good - tight, accurate, and complete to a consistent level of detail (ie: we did not model timber connectors or anchor bolts). Overhead was lower than it would have been if done in 2D CAD.
Overall, the move to BIM was entirely worthwhile. But it took 2+ years and a few hundred unbillable man-hours of training and experimentation for the vanguard to get to break-even. Then it got better and it continues to get better. I have not considered using 2D CAD since the end of year 1. But there is still plenty of room for improvement. Thanks for the soapbox.