Companies–both large and small–must manage increasingly large amounts of data. Within engineering departments, this might include drawings, part files, and lots of other files that represent products. Keeping track of this data is difficult without some type of file management system in place, however, statistics show that the majority of companies continue to manage their data–often inefficiently–without any type of system in place.

Reasons behind why companies don’t implement document management systems, such as PDM or more enterprise-wide PLM systems, vary. Commonly cost is an issue as well as the complexities around implementing these systems. Often engineering departments don’t have the dedicated IT resources to facilitate such implementations, especially at smaller companies.

It’s important for all companies to evaluate the pros and cons of part numbering schemes.

Which part numbering scheme is best?

Part numbering certainly seems like an easy enough task, however, if not done using a consistent methodology, chaos will reign. Turns out nearly every company struggles with defining part numbers.
Without some type of organized system in place, finding the right part files will prove difficult, and time will be wasted. Engineers spend a ridiculous amount of time searching for files, time that could be spent doing actual engineering work.

So the question becomes: should companies mandate use of an intelligent numbering system that embeds important identifying information or go with easy-to-manage generic numbers.

I came across a blog written by Ed Lopategui on the GrabCAD site that seeks to unveil the logic behind a couple of different approaches to part numbering. The author doesn’t seek to crown a winner of this debate, but to help everyone understand the issues and how they will affect your company.

Creation and Data Entry

Generating a new number should be easy, even for the new intern. Nothing is easier than a system-assigned generic number. PDM/PLM systems default to a generic numbering system, but that is neither a recommendation nor a limitation. Most modern systems can accommodate intelligent numbering without heavy customization and/or manual entry. If you choose to embed intelligent information, you are trading the simplicity of creating a part number for a downstream benefit, so weigh that benefit carefully. Consider the part number is often what you start with on new design.

How much will you know at that stage and how certain are you about it? Could that information change in the future? Such considerations will help determine what information, if any, is truly practical to embed in an intelligent number.

Longevity and Legacies

You want your part numbers to last. Intelligent numbering systems tend to break down over time, especially if the intelligence is used for complex categorization. It probably won’t be tomorrow, or next week, but a couple of years from now, someone will likely be staring at a screen and shaking their fist at you because something doesn’t quite fit. The more complicated the system, the higher the likelihood it will break down.

Well-planned systems, however, can last. Before you get too caught up on planning for infinite longevity, keep in mind all of it may come crashing down come a merger or acquisition. Also, don’t forget about the part numbers you already have; you just might be stuck with them.

Readability

Readability is absolutely critical, people need to quickly parse through a large amount of part numbers every day and short-term retention is important. Generic numbering tends to be less readable without some designed structure or variation (i.e. breaking up long series of numbers with letters or dashes at fixed positions). It’s the reason you might remember a telephone number with an area code, but not your license plate, despite a smaller namespace. Intelligent numbers can have readability issues for the very same reasons, or if they just get too long.

Uniqueness

Two parts with the same number is trouble. Some argue that only generic numbers ensure uniqueness, but that’s not really true. You can get the same uniqueness guarantee with the right PDM/PLM system for intelligent numbering. Generic numbers, which tend to be shorter, can actually increase the chance of overlap with respect to mergers and acquisitions or cause confusion with similar supplier part numbers. Nothing ensures uniqueness in this scenario, but the larger the namespace, the lower the chance for a collision. But once again longer part numbers degrade readability.

Interpretation

Every time a part is handled, sorted, searched, or otherwise used an interpretation cost is involved. In other words, it’s the time needed to understand whether you are dealing with the right part. Intelligent part numbers can reduce this interpretation cost, provided the user understands the identification system. In the right conditions, parts can be recognized at a glance. Take caution, however: if the cost of maintaining the intelligent system exceeds the interpretation cost, it’s self-defeating. Generic numbers, on the other hand, can increase interpretation costs, since differences have to be queried in the system. Generic part interpretation can be enhanced with classification systems, but they also add cost.

Balancing all these diverse factors is difficult, because no solution is optimal for every company.

Here are some final tips to help you make prudent decisions:

* Understand your PDM/PLM system part number generation capabilities.
* Understand the limitations of any other systems that interact with your parts.
* Go through every activity that requires interpreting part numbers and understand what system access is available, and how the interfaces work. This will provide a good basis for your interpretation cost.
* Understand how easy/difficult it is for a new employee to interpret a part number

To read the entire blog, click here.