SOLIDWORKS PDM Feedback

PDM
,
1

Hello

We are a company working in the field of special machinery. We easily have machine assemblies exceeding 10,000 parts.
We have 6 people working on solidworks + 5 workstations used for edrawing consultation in our workshop. We work together on the same project.

We are currently using Windows Explorer with these advantages and disadvantages.
Being a small structure, we want to keep this simplicity of operation and not overload ourselves with time-consuming tasks with no added value.
But at the same time, we have problems keeping files locked.
During the course of the study, it is also not uncommon for a modification to be overwritten by an unfortunate backup by an employee.

In short, he comes to ask himself the question of using a PDM.
We want to lock down our data, avoid mistakes, and frame things better by using a PDM.
But we don't want to make our workflow heavier and that's the biggest obstacle.
For example, we don't have to manage different versions. Our design is constantly evolving until delivery, version 1 being the latter.
Another recent aspect, facilitating remote work is an avenue that may interest us. Currently, we have a hard disk backup.

Apart from a salesperson's sales pitch, we would like an opinion from end users.
Do you have any feedback on the use of solidworks PDM in your company?
Is managing the PDM time-consuming every day?
Do you think the added value of a PDM is relevant in our case?

Thank you for your clarifications

2

Hello

Do you have any feedback on the use of Solidworks PDM in your company? It has been in place for 6 years with us (with two remote sites and 2 independent databases: 1 per site but the other site can access it if it wants). Our largest files are in the 2000 components (200 unique files)


Is managing the PDM time-consuming every day? It is the validation of the files that is time-consuming. But if you do an ultra-basic and almost empty workflow, without validation it's not too time-consuming (it's still longer than nothing). We have a main workflow that allows us to make minor changes without going through an approver for example: this kind of transition takes a little time but not that much. After that, if you validate a machine of 10000 parts and you want all the components to be validated at the same time, it will grind severely I think. There is therefore a compromise to be made between the "framing" of users and the ease/cumbersomeness of use. 


Do you think the added value of a PDM is relevant in our case? PDM makes remote work much easier (since everyone works with their workstation's local files). In these times of Covid it's not bad (I'm working from home at the moment for example). Having already used external usb / dd sticks I know that it's very quickly a hassle when you work with several people on a project. The big plus of PDM is that it allows you to keep a history of everything (so it is quite possible to reopen an assembly in its state of 3 months ago: it makes it easier to understand any plan problems). Keeping all the history also allows you to limit control over users (in practice you can't lose a file even if someone has done something stupid: it's always possible to go and get a previous version)

There are PDM viewer licenses for workshop users (cheaper).

The thing to take into account is also the intrinsic cumbersomeness of the system. You can outsource the implementation of the PDM, but it's still better to have someone smart enough to be more or less autonomous on it (to manage rights, create and delete users, make minor changes to workflows, etc.).

EPDM also allows you to automate certain tasks: creation of the pdf (or export files) automatically after validation for example. This saves clicks that you have previously lost with workflow management.

As EPDM is backed by a database, it is also very easy to find cases of use of parts (unless you use virtual sub-assemblies: EPDM does not handle this very well). This can allow you, for example, to standardize certain parts and reuse them safely in several machines (for having worked 6 months in a special machine, the company where I was spent its time redesigning/renaming identical timing washers because each part of a machine had the number of the machine in its name...).

For us, the main reason for switching to EPDM was the collaboration between our 2 sites.

If you're ISO 9001, auditors love it: all changes and approvals are tracked.

In the end, EPDM makes it easier and more secure to collaborate, but this has a direct economic cost (licenses are not cheap, whether in purchase or maintenance) as well as generally a certain cumbersomeness that can penalize the work (if the workflow is complicated).

 

 

3 Likes
3

Hello

Thank you for your answer.
We suspected that this additional management adds a certain cumbersomeness. In the end, it is a necessary evil to guarantee data security.

When you talk about keeping the history of changes: are we talking about changes between each record in the PDM or between each validation process? In our country, a study can take 3 months to validate it. We need to have back-ups during these 3 months.

For the implementation in your sites, how did your process go, should you expect a downtime of the studies during it?
Do we need a special treatment of the history to integrate it into the system?
Or is it better not to go back to the history and feed the system just with the new studies? (Even if it means transferring certain important cases on a case-by-case basis.)

4

Hello

I'm going to provide answers on the last post since I'm in tune with Froussel's answer.

For the change history, it is because you can check in intermediate versions before arriving at the final version. This allows you to go back if necessary and freeze versions launched in analysis for example or in prototype...

As far as the implementation is concerned, there will indeed be a shutdown of around 3 to 4 days if my memory is correct (optimizeable by transferring the data during the weekend) but you can take advantage of this time to train users in the use of PDM. You can also set up a workspace on specific data to identify changes between the transfer of PDM servers and the opening of access (this will allow you to reload this data if you do not use library-like components such as screws in these working files)

For history, it all depends on the quality of your data. If your files are properly managed at the moment (no CAD in the premises of some users), your history will go back perfectly.

After that, it's more a question of "forcing" the use of this tool. If your users find the tool too heavy, in the event that CAD is available elsewhere, they will be tempted to work the old way (unless you lock everything in read-only) and you will lose the added value of PDM.

It really depends on the habits of your 6 users and whether or not they are closed to changing their working methods.

5

We all already work on the server, no one works locally. This is already a good point.
We are a young team and rather open in the offices, I am not too worried about that. Maybe rather on the workshop side where it may be more difficult.

In terms of history, we have one backup every evening over a rolling week after one backup per week over a rolling month and then one backup per month. But I imagine that we can set up a routine to keep a history even without having validation?

When you talk about this tool facilitating the standardization of parts, what use do you have in mind?
We have a standard off-the-shelf part library and a standard part library of our manufacture. These 2 libraries are read-only locked. Apart from the libraries, each room bears the name of the case.
In your case, in one case you end up with pieces from other cases?

Overall, I imagine that we have to change our working methods. We probably have to adapt the way our files are named, filled in, organized etc. For this part, I think Visiativ is a good help. How did this phase of reflection go for you?

6

We carried out an audit with Cadware at the time (now an integral part of the Visiativ group) on our internal processes and naming rules. In the end, we haven't changed a lot in the way we name our CAOs so as not to lose the habits of users, but everything is possible depending on the configuration we want to have (prohibition of duplicate names, unique reference, etc.).

For history, there is no need for a validation status if you do not add a rule for purging versions not backed by revisions (vault configuration). The vault is based on a SQL database and a file server, each archived version corresponds to a file stored on the server.

Backups are therefore to be made on the server side (SQL database and data server) and you will have to be careful with archiving because in the event of a PC crash (SSD/HDD HS) the data of the version being modified would be lost (maintenance of lower versions).

7

The switch to EPDM depends on the quality of your current data. If you have no (or few) duplicate files (same name but not stored in the same place), re-importing your previous stuff is possible.

The interest of a change of this magnitude is to also be able to make the necessary changes to the tree structure (tidying up if it was not the case before).

I know that some resellers have tools to analyze your current tree: this allows you to judge the quality of your data and tell you if it is better to forget about it or if you can import it.

In any case, the whole company must switch to EPDM (and therefore prevent users from working according to their old habits).

Having the reseller on site during the implementation seems essential to me. It may even be necessary to separate the part of creating the workflow(s) and configuration from the part of switching to EPDM + user training. Because when creating workflows and data maps, there may be some questions that you won't be answered immediately and that will require you to think about.

It's best to take a little time when setting up the system, because once your data is in it and your users have started working on it, it may be more (or even much more) complicated to scale.

 

1 Like