I'm trying to find statistics on the average number of drawing revisions for projects. Difficult question with lots of variables... I know... but I haven't been successful in finding anything remotely close and wondering if somebody has come across this or can point in the right direction.
it's an absurd question. on most projects the design is a process of continuous incomplete revision. nothing is done until it's built. what are you trying to figure out with this hypothetical information?
It's an absurd comment, but thank you for asking. It's for a market study. We all know it's a continuous process of revisions through construction, which is why I can count the # of submissions on all of my projects through C.O., others should be able to count theirs, and why I believe at some point somebody must have compiled some statistics on it.
This is difficult to control or reduce but conducting a weekly design workshop and continuously demanding a user-design brief (client side) would assist in completion of package.
well depends and depends.....i have a titleblocks with 15+ issues or just 1. too many factors to consider. in the design process revisions are allowed to a degree but after i switch to CD or filing its all additional services and quite often if the change is inconsequential to sign-offs I do not bother drawing it.
There are no averages nor statistics for "revisions" as projects vary wildly.
Ideally you can have some revisions during Schematic Design, even less at Design Development and if you are very disciplined and great at what you do, at the end of 100% DD Phase completion all design/consultants/utilities wrinkles have been ironed out and quite minor revisions should be expected as a consequence.
Finally at Plan Check, further minor revisions are typically req'd to comply with code compliance. TYP. U.N.O.
Revisions are changes made after a design is sent to bid and or permit. Is that what we agree on as to what we are talking about?
This is also a thorny question as many people look at revisions under the veil of errors and incomplete information as much as it is finicky clients that keep changing their mind and circumstances beyond anyone's control. Sometimes these are just part of the process but sometimes they can have impacts on time and cost. Significant and insignificant revisions happen all the time.
For example we had a project that was going to be precast structure mostly, then a tornado wiped out the precast facility so we switched in a revision to steel because we could not wait for the precast-er to rebuild operations but still used them for exterior panels once they got back on line.
It is very rare that there are no revisions on a project sometimes the cause and impact can be interesting.
I am interested only in construction set revisions. I'm not looking for progress or design iterations. I knew it was a hard question and of course nobody will be able to give a finite answer... I'm just curious if anybody had come across statistics. I think it's safe to say everybody should be able to (at the end of a job) count up the number of drawing revision submissions made... so I don't think it's an "impossible" answer as some of you seem to think it is. I'd guess my average would probably be 6-8 average drawing revision per project, including Arch, MPE & Structural, but I'm looking for hard statistics.
I think everyone is pointing out that there aren't any good statistics because the variation is so wide - as much as analytics are in vogue today, I'm not entirely sure what you're getting at. I can say at the low end it's 2~3 (coordination done as coordination set phase, hence only field related changes, which can be minimized with a thorough survey) at the high end I've seen over 10: 4~5 'final construction sets' issued prior to revisions (they all had some wording of 'final' after the 100% CD was issued).
1. depending on the scale of changes and type of project changes may be issued as sketches not full revisions.
2. number of coordination related revisions can depend greatly depending on the design team, the staff at each of those consultants, nature of project (client, funding, changes in cost due to escalation, etc).
It's not an 'impossible' answer, it's just one that makes little sense to track. Most firms quickly figure out which team members and clients are liabilities in terms of the need for revisions and avoid them.
PeterN: revisions that are somewhat common but annoying for me: electrical duct banks (because they often aren't located or size called out on surveys) and historic artifacts (unmarked graves included).
Too many for me.. I've revise one of my shopping mall schematic design for more than 5 times in 2 years.. I mean the whole package, not considering small revision such as fire stair relocation, etc etc..
Then it will always be revision on tender doc, construction doc until the soft opening :P
ctalley5: as you are learning here, there are very few reliable sources for the kind of data you seek. However, you might want to research design firms that have adopted, and been certified as satisfying, a formal "Qualify Management System" (QMS) such as ISO 9001. Such firms are rare in our profession, but they do exist.
For a QMS to actually be meaningful, drawing changes must be documented, along with the reason for the change and any sort of "corrective action" that was implemented in an effort to prevent that particular sort of revision from occurring again in the future. Such systems typically track the number of changes / errors being encountered in an effort to verity that overall production quality is improving over time.
Change management fail or client management fail...I'm unfortunately working on a project currently on its 17th addendum. In this case the client won't(can't) stop changing their mind.
The key is, get paid to put it in and get paid to take it out.
it would be hard to learn anything only by looking at the number of cd re-issues without knowing the context and scale of each project. and it would be worth tracking not just whole set re-issues, but how much content actually changed . was it 30 plan sheets reissued because one gridline moved to accomodate unexpected buried obstructions, or 3 detail sheets revised because the original design for a special facade was dysfunctional? the number of sheets doesn't necessarily tell you how 'severe' a revision is.
in my experience the projects with exceptionally large numbers of set reissues are due to clients with competing internal demands that are never fully resolved, long lead proejcts which result in market changes for materials and commercial spaces, and projects which involve politcal stakeholders who tend to mess with things on their own whims. to a lesser extant some changes are due to rushed construction schedules, done with the knowledge the initial set isn't going to be final. design error was never sole cause to re-issue a set, though some revisions are an opportunity to address what no one noticed yet.
the most egregious design mistakes don't result in revisions; they lead to replacing the architect - so really big mistakes may not show up in any statistic.
Mar 30, 16 12:00 pm ·
·
Block this user
Are you sure you want to block this user and hide all related comments throughout the site?
Archinect
This is your first comment on Archinect. Your comment will be visible once approved.
Average number of revisions per project
I'm trying to find statistics on the average number of drawing revisions for projects. Difficult question with lots of variables... I know... but I haven't been successful in finding anything remotely close and wondering if somebody has come across this or can point in the right direction.
Thanks
three
it's an absurd question. on most projects the design is a process of continuous incomplete revision. nothing is done until it's built. what are you trying to figure out with this hypothetical information?
It's an absurd comment, but thank you for asking. It's for a market study. We all know it's a continuous process of revisions through construction, which is why I can count the # of submissions on all of my projects through C.O., others should be able to count theirs, and why I believe at some point somebody must have compiled some statistics on it.
This is difficult to control or reduce but conducting a weekly design workshop and continuously demanding a user-design brief (client side) would assist in completion of package.
Leandro, bullshit. The more hands-off my retail clients are, the better the project ends up being.
well depends and depends.....i have a titleblocks with 15+ issues or just 1. too many factors to consider. in the design process revisions are allowed to a degree but after i switch to CD or filing its all additional services and quite often if the change is inconsequential to sign-offs I do not bother drawing it.
I note that you comment "number of drawing revisions". You can change the drawings because:
-there are errors
-they do not coordinate with other drawing sets, i.e. mechanical, structural, electrical
-the design is evolving during schematic design through detailed design via internal decisions or ones driven from the client.
-revisions are requested in the value engineering phase through the use of a quantity surveyor or a general contractor.
-changes must occur because issues arise during the construction phase which were discovered on site.
Which changes would you like to pinpoint, or do you intend to count them all?
There are no averages nor statistics for "revisions" as projects vary wildly.
Ideally you can have some revisions during Schematic Design, even less at Design Development and if you are very disciplined and great at what you do, at the end of 100% DD Phase completion all design/consultants/utilities wrinkles have been ironed out and quite minor revisions should be expected as a consequence.
Finally at Plan Check, further minor revisions are typically req'd to comply with code compliance. TYP. U.N.O.
Revisions are changes made after a design is sent to bid and or permit. Is that what we agree on as to what we are talking about?
This is also a thorny question as many people look at revisions under the veil of errors and incomplete information as much as it is finicky clients that keep changing their mind and circumstances beyond anyone's control. Sometimes these are just part of the process but sometimes they can have impacts on time and cost. Significant and insignificant revisions happen all the time.
For example we had a project that was going to be precast structure mostly, then a tornado wiped out the precast facility so we switched in a revision to steel because we could not wait for the precast-er to rebuild operations but still used them for exterior panels once they got back on line.
It is very rare that there are no revisions on a project sometimes the cause and impact can be interesting.
what was your weirdest Revision?
Over and OUT
Peter N
I am interested only in construction set revisions. I'm not looking for progress or design iterations. I knew it was a hard question and of course nobody will be able to give a finite answer... I'm just curious if anybody had come across statistics. I think it's safe to say everybody should be able to (at the end of a job) count up the number of drawing revision submissions made... so I don't think it's an "impossible" answer as some of you seem to think it is. I'd guess my average would probably be 6-8 average drawing revision per project, including Arch, MPE & Structural, but I'm looking for hard statistics.
I think everyone is pointing out that there aren't any good statistics because the variation is so wide - as much as analytics are in vogue today, I'm not entirely sure what you're getting at. I can say at the low end it's 2~3 (coordination done as coordination set phase, hence only field related changes, which can be minimized with a thorough survey) at the high end I've seen over 10: 4~5 'final construction sets' issued prior to revisions (they all had some wording of 'final' after the 100% CD was issued).
1. depending on the scale of changes and type of project changes may be issued as sketches not full revisions.
2. number of coordination related revisions can depend greatly depending on the design team, the staff at each of those consultants, nature of project (client, funding, changes in cost due to escalation, etc).
It's not an 'impossible' answer, it's just one that makes little sense to track. Most firms quickly figure out which team members and clients are liabilities in terms of the need for revisions and avoid them.
PeterN: revisions that are somewhat common but annoying for me: electrical duct banks (because they often aren't located or size called out on surveys) and historic artifacts (unmarked graves included).
Maybe you should track non-owner initiated change orders as a percent instead? Would actually be useful data.
Too many for me.. I've revise one of my shopping mall schematic design for more than 5 times in 2 years.. I mean the whole package, not considering small revision such as fire stair relocation, etc etc..
Then it will always be revision on tender doc, construction doc until the soft opening :P
ctalley5: as you are learning here, there are very few reliable sources for the kind of data you seek. However, you might want to research design firms that have adopted, and been certified as satisfying, a formal "Qualify Management System" (QMS) such as ISO 9001. Such firms are rare in our profession, but they do exist.
For a QMS to actually be meaningful, drawing changes must be documented, along with the reason for the change and any sort of "corrective action" that was implemented in an effort to prevent that particular sort of revision from occurring again in the future. Such systems typically track the number of changes / errors being encountered in an effort to verity that overall production quality is improving over time.
Good luck.
Change management fail or client management fail...I'm unfortunately working on a project currently on its 17th addendum. In this case the client won't(can't) stop changing their mind.
The key is, get paid to put it in and get paid to take it out.
it would be hard to learn anything only by looking at the number of cd re-issues without knowing the context and scale of each project. and it would be worth tracking not just whole set re-issues, but how much content actually changed . was it 30 plan sheets reissued because one gridline moved to accomodate unexpected buried obstructions, or 3 detail sheets revised because the original design for a special facade was dysfunctional? the number of sheets doesn't necessarily tell you how 'severe' a revision is.
in my experience the projects with exceptionally large numbers of set reissues are due to clients with competing internal demands that are never fully resolved, long lead proejcts which result in market changes for materials and commercial spaces, and projects which involve politcal stakeholders who tend to mess with things on their own whims. to a lesser extant some changes are due to rushed construction schedules, done with the knowledge the initial set isn't going to be final. design error was never sole cause to re-issue a set, though some revisions are an opportunity to address what no one noticed yet.
the most egregious design mistakes don't result in revisions; they lead to replacing the architect - so really big mistakes may not show up in any statistic.
Block this user
Are you sure you want to block this user and hide all related comments throughout the site?
Archinect
This is your first comment on Archinect. Your comment will be visible once approved.