CAD Dept. Management Challenge!! (Productivity)

[Hans_2011]

Hello Hans,

You've touched upon a subject that is very important to me. But first let me comment on something you said in your post. The general gist of it was that you are creating a 3d model from your 2d design. It is my opinion that this approach is exactly backwards from a more efficient approach of creating the 2d details (or layouts, etc) from the 3d model. How did I form this opinion? Hopefully It will become evident.

So, to your main theme, you are interested in determining a method whereby you can estimate more accurately the amount of time it takes to produce a drawing (or design, or project).

First and foremost, do you know accurately how much time it had previously taken to produce a particular drawing? You must establish a benchmark, and really a very gross estimate isn't good enough.

Next, do you know how long it takes you to produce a particular type of drawing? Do you know how long it takes Hans Johnson. to produce a similar drawing? Do you know if your efficiency is greater or less than his? Are the drawings in question worked on by both you and him? Is one producer 's final drawing a higher quality than the other and conforms to the standards in place?

If you do not know the answers to these (and many more) question then you need to put into place the means to begin answering them. You need to begin monitoring, logging and examining the logs. You might want to start by trying my CadTempo software http://www.cadtempo.com for a 30 day trial.

Ok, how did I form my opinion about going from 3d to 2d? hopefully my departure from your question will help answer some of your other questions.

I started my business 20 years ago. Prior to that I had been doing the same type of work on the drawing board and had kept meticulous records of my time involvement. I do machine design and each machine will have a number of compontents whether they are designed, detailed, and manufactured or are commercially available. By knowing the component count I could determine how long an average machine of xxx component type took to design or "layout", then detail, then check.

During start up I decided to do all my work in AutoCAD because it was highly customizable and also offered AutoLISP as a programming language. I discovered that AutoCAD's TIME command was essentially useless. When you save one drawing into another name the time would carry along. Also if you left a drawing open the timer would just keep ticking along. Not a very good way of determining how much time it took to produce a drawing. So the first order was to create a lisp program that would track my time more accurately.

I learned it took longer to produce a design than manually drafting did, much of it having to do with the learning curve. After some menu tweaking, some custom lisps etc the time came down to a little less than board drafting. I knew I could get it down further so I set a goal for myself and began determining what needed to be done. Looking at my logs I could determine if things appeared out of the ordinary from my previous experience. And it occurred to me. Why am I drafting most everything six times??? Yep, that's right six times - during layout: front, top, side...during detail: front top side. Why not draw it once in 3d then extract the drawings and views as needed.

Any long term AutoCAD user knows that the earlier releases of AutoCAD did not lend themselves to convenient 3d work so I had to create some 3d lisp routines and menus and macros to speed things up. It worked! My design and detail time started improving dramatically, I'll not bore you with details but I continued to improve my methods and techniques and a lot of different things. Every change resulted in moving my time up (bad, so let's not do that) or down (good, ok maybe I can do that even better).

It is not a task to be taken lightly, it takes persistence and determination to arrive at your goal but it starts out by benchmarking.

Dear Duhvinci,

We design in 2D first since it takes fewer time than drawing in 3D at first.
Clients require a specific timeline for presentation of General Arrangement 2D Drawings.

In any case, you have made an important point here.
We use AutoCAD for 2D and Tekla Xsteel for 3D. But most probably we can use Tekla for 2D design as well... I am not sure if we are using the AutoCAD 2D finished drawings to import to the 3D model and start from there, so I will have to check.

Thank you for your posts...

[Hans_2011]

I haven't read all of the responses on here, so forgive me if I repeat anything already said.

To address the most recent points, I personally can attest to my frustration over these issues during my own career history. I am referring to the project timeline part mostly.

In my experience, and I have been doing this job now for more than 12 years. Currently, I am at the point where I've had enough of it. There is a culture in this industry (at least in my country) where the drafters are not only often required to do overtime, but it is an expectation. Now I wouldn't have all that much problem with overtime, except it's not ever paid for, or compensated. Several companies I have worked for have a "no time in lieu, no overtime pay" policy. In fact one company I worked for, the project manager promised me time in lieu to finish off a project to meet issue deadlines. Myself and another drafter came in, worked 8 hours saturday and sunday, and I then followed that up with a 12 hour monday. I was thinking sweet, almost 3 days off I can take a long weekend and catch up with my girlfriend, but on the tuesday myself and other drafter were taken aside and told that the company had a "no time in lieu or overtime" policy. That's great, thanks for promising something then taking it away after we've done the work for you. How this is different to stealing I'm not sure. Needless to say, this resulted in both of us quitting that work not long after.

My point really is, that its very hard to stomach the fact that my life, my time, and my effort are EXPECTED to be given for free to fulfill the unrealistic promises that someone else made without even consulting the person involved on how long they estimate tasks should take. Essentially they are writing cheques that I have to fund. Not only does this cause resentment, there is absolutely no or very low motivation from day to day pretty much caused by over working, burnout is probably the best way to put it. Drafters are treated, in general, as something you can just use up, burn out, and replace.

Without going on too much with sour grapes, I am hoping you've got the idea that this builds resentment in your workforce. Compensate your staff for overtimes, time in lieu at the very minimum. Keep them VERY aware of timelines and consult them regularly regarding the timelines, and furthermore take their advice when they tell you something wont be done in time, or is unrealistic. Meaning allocate more resources, dont just expect someone to stay all weekend to pick up the slack. If they can't hit a deadline it doesnt mean they arent working hard, maybe it means unrealistic timelines have been set.

In a carrot on a stick reference, the carrot is much more effective when used as an incentive rather than the stick being used to beat, and the carrot inserted in the other end. I can tell you right now the whip is not a good motivational tool, so the timelines/overtimes issue is way more important than you might think.

Dear Karalon,

Very interesting to hear the other side of the story, at least from your experience.
I understand this may be the feeling of my employees as well, but if you think about it, the problem is the lack of communication to be able to arrive to a fair measurement and reward system that would benefit both parties.

I am in the problem of not being able to find the solution to put the right measurement of productivity in place to reward the good performers instead of rewarding everyone for overtime which doesn't pay off to the company as an increased work output.

Thank you for your post.

[Hans_2011]

This is where one of our metrics comes in to play - trend reports. Whenever somebody wants to make a major change to a project, they submit a trend report containing whats getting changed, what the importance level is, what impact it will have on the schedule, who is requesting the change, supporting documentation, and so on. Trends get reviewed by managers and/or the client who approve or reject the changes, based on consulting the client, experience, and just plain talking with the design group. This way if the schedule starts expanding, issue dates slip, etc. there is a paper trail of where the extra time is going and agreements of whats been changed. If there is no reason for missing scheduled issue dates then those responsible aren't qualified and get re-assigned to something more suited to their level of ability - which sometimes means being shown the door. After particularly bad projects there is also a post-mortem to find out why and how it went wrong.

Dear dgorsman,

I agree to everything you mention.
Progress reports though, at least in our case, rely on what the drafters say the status is and we don't have very good technology to measure and take reports.
We need to improve a lot in terms of processes to arrive to what you mention.

Something I have not mentioned yet, is that we are a very small company, and we don't have a lot of resources to dedicate exclusively to some of these tasks.
I need to prove that if we go and invest in this, we will be able to see results in the bottom line.

Thank you for your post.

[dgorsman]

We aren't that big, either, and trend reporting is zero extra cost - its all part of project managment. The work of approving and reviewing trends is the part of the managers, and entering/tracking the information is delegated to their assistants. Trend reports are nothing to put together, in some cases its just a memo format.


If you are looking for a magic metric, you aren't going to find one. If you can't rely on the designers to say what the status is (like "I'll be done this drawing this afternoon" or "These revisions will take a week") then look at the drawings yourself and estimate how long it should take. If it takes longer, move your estimates up. If it takes less time, move them down. "All our projects are different" doesn' wash; a company that small can't do *everything*, you are limited to a particular area of expertise. While the projects may differ the generalities are the same, and the time required will be approximately the same as well.

[Liamnacuac]

I disagree. projects at small companies can vary greatly. If the company has been doing very similar projects for quite some time, the relative speed of performance will be high as certain operations become routine. However, when the project types change or the requirements from the customers change, then your performance will degrade. What is needed is experience with similar types of projects in which they can gauge their time. If they have had considerable turn over, then estimating different drafting capabilites is also difficult to gauge.


If you are looking for a magic metric, you aren't going to find one. If you can't rely on the designers to say what the status is (like "I'll be done this drawing this afternoon" or "These revisions will take a week") then look at the drawings yourself and estimate how long it should take. If it takes longer, move your estimates up. If it takes less time, move them down. "All our projects are different" doesn' wash; a company that small can't do *everything*, you are limited to a particular area of expertise. While the projects may differ the generalities are the same, and the time required will be approximately the same as well.[/QUOTE]

[Hans_2011]

Hi All,

I thank you everyone for the replies.
It seems this post is dying long before I get answers to the first question so I would like to summarize...

I think the most useful response was the one suggesting to use CADTempo to implement and measure man-hours as an INPUT to the egineering design process.
My main problem is to quantify the Scope of Work.

Answers like: "You can't determine this...", "You have no other option to rely on what the drafters tell you everytime...", etc... are not useful in managing better. Doing as suggested would mean to keep inefficiencies forever and in the end, keep the department to be managed by the drafters and not the managers. I strongly believe there is a way to measure the scope of work correctly and together with solutions such as CADTempo, it will be possible to determine productivity.


Just to prove my point, and to give hope to all of the people who read this, I can tell you that after some research, I have found some extremely interesting papers talking exactly about what I need:

1. "Quantifying Engineering Project Scope for Productivity Modelling", Journal of Construction Engineering and Management ASCE, March 2005 - Lingguang Song, and Simaan AbouRizk.
2. "Measuring and Estimating Steel Drafting Productivity", Construction Research Congress, 2003 - Lingguang Song, Michael Allouche and Simaan AbouRizk.

The one who is really looking for a solution to this should buy these papers and read them. In the end the idea is not far from my original concept (before in this post) of determining how much it takes to draw a cube compared to a cube with a hole in the middle.
Check them out!

I am currently gathering internal information of past projects to put these methods to test...
Thank you Mr. Lingguang Song and colleagues for your work!

[Hans_2011]

Hello Hans,

You've touched upon a subject that is very important to me. But first let me comment on something you said in your post. The general gist of it was that you are creating a 3d model from your 2d design. It is my opinion that this approach is exactly backwards from a more efficient approach of creating the 2d details (or layouts, etc) from the 3d model. How did I form this opinion? Hopefully It will become evident.

So, to your main theme, you are interested in determining a method whereby you can estimate more accurately the amount of time it takes to produce a drawing (or design, or project).

First and foremost, do you know accurately how much time it had previously taken to produce a particular drawing? You must establish a benchmark, and really a very gross estimate isn't good enough.

Next, do you know how long it takes you to produce a particular type of drawing? Do you know how long it takes Hans Johnson. to produce a similar drawing? Do you know if your efficiency is greater or less than his? Are the drawings in question worked on by both you and him? Is one producer 's final drawing a higher quality than the other and conforms to the standards in place?

If you do not know the answers to these (and many more) question then you need to put into place the means to begin answering them. You need to begin monitoring, logging and examining the logs. You might want to start by trying my CadTempo software http://www.cadtempo.com for a 30 day trial.

Ok, how did I form my opinion about going from 3d to 2d? hopefully my departure from your question will help answer some of your other questions.

I started my business 20 years ago. Prior to that I had been doing the same type of work on the drawing board and had kept meticulous records of my time involvement. I do machine design and each machine will have a number of compontents whether they are designed, detailed, and manufactured or are commercially available. By knowing the component count I could determine how long an average machine of xxx component type took to design or "layout", then detail, then check.

During start up I decided to do all my work in AutoCAD because it was highly customizable and also offered AutoLISP as a programming language. I discovered that AutoCAD's TIME command was essentially useless. When you save one drawing into another name the time would carry along. Also if you left a drawing open the timer would just keep ticking along. Not a very good way of determining how much time it took to produce a drawing. So the first order was to create a lisp program that would track my time more accurately.

I learned it took longer to produce a design than manually drafting did, much of it having to do with the learning curve. After some menu tweaking, some custom lisps etc the time came down to a little less than board drafting. I knew I could get it down further so I set a goal for myself and began determining what needed to be done. Looking at my logs I could determine if things appeared out of the ordinary from my previous experience. And it occurred to me. Why am I drafting most everything six times??? Yep, that's right six times - during layout: front, top, side...during detail: front top side. Why not draw it once in 3d then extract the drawings and views as needed.

Any long term AutoCAD user knows that the earlier releases of AutoCAD did not lend themselves to convenient 3d work so I had to create some 3d lisp routines and menus and macros to speed things up. It worked! My design and detail time started improving dramatically, I'll not bore you with details but I continued to improve my methods and techniques and a lot of different things. Every change resulted in moving my time up (bad, so let's not do that) or down (good, ok maybe I can do that even better).

It is not a task to be taken lightly, it takes persistence and determination to arrive at your goal but it starts out by benchmarking.

Dear Duhvinci,

By the way, I forgot to comment about your CADTempo solution. This is something I am craving for for long...
I have checked your website of the software but I am finding difficult to understand all thecharacteristics of the software and what it actually measure and what it doesn't.
Do you have a more extensive manual for the program? I am more than willing to buy this if I understand it better.
Keep in mind I need it for controlling the work of app. 15 drafters on a network. I have lots of questions but this ones to start with:
1- What actions in AutoCAD does it track? Which ones doesn't? Example: Does it log time when the computer is idling at lunch time with a dwg open? Does it log activity when a dwg is open but activity is happening in Internet Explorer? I need to understand exactly what activities is tracking and what it doesnt.
2- How can I implement in a network without letting users get access to the log files and software? DWG files are stored in a shared network drive to which everyone has access. I can put the CADTempo files and logs in a specific folder with restricted access only to managers for example, but from what I read in your website there would be problems if I do this since the software users need write access to the log files? Please explain.

Maybe we should take this out of this post? Can you see my email from my profile?
Regards.

[duhvinci]

Hello Hans,

I'm pleased to hear this is something you find may be helpful to your cause. I would like to work with you in any way I can.

Here is a synopsis of the functioning of CadTempo. Beyond that you are correct, we should take it to a different means of communication after all this is a forum for Autodesk products in general and management in particular. May I suggest the support forum for CadTempo that I've set up? You can visit here: http://engds.com/supportbb/index.php I will also e-mail you.

CadTempo has 4 timing functions:

Session timer: This timer logs the amount of time monitored applications are open. This timer can serve as a timesheet recorder.

Document timer: This timer logs the amount of time a monitored document (.dwg, .rvt, .ipt, .xls, etc) is open, in focus, and being worked on. This timer records the elapsed time of the document of each individual user. The timer also records the exact edit time of the document and users based on an adjustable idle timer setting. With these two measurements an efficiency rate can be established. I've found for my own personal usage my efficiency can be as high as 100% for things like detail or revision work to 85-90% for design work, and down to 50% while needing to do research or examine reference material. You would be able to establish norms for various members of the team.

Task Timer: This timer (if activated) automatically prompts the user for a task name when a session is started. It remains in the Windows task bar and can be finished as needed, then put to sleep and re-prompt.

Activity Timer: Although named "Activity Timer" it counterintuitively is monitoring inactivity. When this timer is initiated (following a set idle period) when the adjustable "inactive" period is reached a prompt is issued to the user to enter an activity such as Lunch, or Meeting, or Customer Visit. The purpose of this timer is to provide the user a means to explain the reason for an extended idle time. It is issued after the inactive period because it often times cannot be anticipated and might not have been started at the beginning of the inactive period, such as an interruption.

The network installation of CadTempo installs the programs (logger and viewer) on a network shared folder. You can select separate locations for each. When installing for a user you select either the viewer or the logger. Each only provides a shortcut to the shared program location. In the case of the logger a registry entry is also made. Regarding the logs themselves, the document logs are created in a hidden fashion and are not visible to any user. You are correct, write permissions are required for the logs however they are encrypted. If it is a problem that some user becomes devious I would consider implementing counter measures for the program.

Well I hope this provides you with the information about CadTempo you seek. You would think that something as simple as keeping track of time is a simple endeavor, it actually has some rather complex considerations to be accounted for an I believe CadTempo addresses those issues.

I look forward to working with you. Please keep an eye out for my e-mail

[Brian Myers]

Am I the only one that finds this conversation a bit disturbing?

There are two main options:

A.) The employees are doing their job.
B.) The employees are not doing their job.

We also have an issue where management either doesn't know what it takes to get the work done or management does understand and is currently letting their employees get away with it.

So a decision is being made to create a metric that will hold employees accountable for the amount of work they get done. Apparently this is a metric being created by individuals that actually have no idea how long it should take to get the job done otherwise this metric wouldn't be required... if they did understand then a deadline would be set, they would need to get it done by the deadline or have shown commendable level of work toward achieving it. Anyway, since this doesn't apply, it appears we're developing a benchmark with the work of individuals that are being accused of being intentionally slow so that the benchmark itself will not be accurate.

The sad thing is that these employees should be getting paid salary. That way if they intentionally worked slow they would be wasting their own time, not the company's time. If they were still obviously wasting the company's time then they should be replaced, with knowledgeable management it doesn't take a metric to realize an employee just isn't good enough.

In short, it sounds like these metrics are being developed so management can use them as metrics against their own employees. They plan on either using these metrics as a weapon to keep employees on task or as a defense when it comes time to fire them. So there may be as many management issues involved as employee ones.

Of course, the alternative is that the Engineering management DOES understand but their employees are in conflict with management and the owners just don't know who to believe. So these metrics are being developed to prove to ownership these employees need to change the way they do things or be replaced. Which is also a sad situation because this means the owners don't trust the judgement of their own management team.

Now I'm depressed...

[BlackBox]

Am I the only one that finds this conversation a bit disturbing?

1+

Without someone IN the CAD group who knows what work being done actually contributes to the project's completion, your metric is meaningless.

A CAD user can run commands all day long, drawing lines, being productive, and then delete all their work before closing their AutoCAD session... 0% productivity, 100% utilization.

Software is good for many, many things (nothing against CadTempo)... however, resolving this issue in an equitable, sustainable manor, requires human beings working together as professionals. Management needs to know what it is that they're managing, every bit as much as the CAD users need to understand (and adhere) to an comprehensive CAD standard.