Over the years, the Convergence Data team has spoken with many organizations that leverage a PLM solution. As our team wraps up 2022, we reflected on the many excellent interactions and dialogues that we had on the topic of parts classification and decided to share them with the broader community.
Many end users aren't driving as much value out of the platform as they could or should be. Why? The teams haven't embraced a "parts-centric" strategy to influence and govern how their parts are created, classified, or managed over the longer term. This matters for the future state of any PLM initiative, particularly when users look to drive deeper capabilities out of the solution.
Here are eight of the most common reasons we hear from PLM users (across many industries and PLM solutions) on how they are struggling in this area, along with some recommendations that may help:
PLM is only used for file storage. Are you using your PLM solution as a CAD vault? That’s an expensive solution for just storing CAD files! Many organizations fall into the pattern of underutilizing their PLM solution and instead, rely on external workarounds such as homegrown systems or manually managing their parts data in spreadsheets. Many of our customers opt to accelerate their PLM modernization efforts by first establishing a central and foundational classification structure for their parts data, enriching and validating that data, and then, loading it into a target PLM solution. After all, a PLM solution’s functionality can only be as robust as the underlying parts data and attribution that supports it.
Engineers can’t find the parts that they need. Findability is one of the challenges that customers most frequently share with us. Engineers, by their very nature, like to build and create things. When part findability proves challenging and time-consuming, it can be more expeditious for an engineer to create a new part on the fly rather than sift around in the PLM or other siloed data sources to locate a part and its corresponding attribute data. This often leads to a proliferation of duplicate parts in the PLM landscape. While having duplicate parts may not seem like a big deal, there is a longer-term cost to that proliferation. The lifecycle costs of a part (factoring in elements such as material, handling, design, and procurement costs) can add up significantly over time.
Everything is stored in a spreadsheet. Spreadsheets are the most commonly used data solution across industries and domains. Yet, while spreadsheets may seem like a relatively innocuous and inexpensive approach to managing data, it becomes quickly burdensome given the manual effort needed to maintain them over time and pass them back and forth among team members. (As many of our customers know, that “back and forth” takes valuable cycles away from other key objectives and activities.) What’s more, this approach is prone to errors, and, ultimately, unsustainable over time, particularly for teams that need to incorporate the legacy data from other companies they acquire. Spreadsheets may work as an interim solution, but many teams recognize that it's not a scalable one. In addition to establishing a central point for parts data classification, we also recommend focusing on change management to drive and communicate a more cohesive parts data strategy across the team. Ultimately, having a central source of truth for your parts data will get your team out of spreadsheets and working at scale within the same solution.
Parts aren’t in the PLM – they’re everywhere! In short, you can’t manage or report on it if you can’t find or organize it! If the parts aren’t ultimately stored in the PLM, end users will not be able to build and create their Bills of Materials (BOMs). This is a common challenge and can be especially pronounced for organizations that acquire their parts through acquisitions. Those company acquisitions often come with years of legacy parts data scattered across various organizational siloes. The key remedy to control this sprawl is to develop a foundational classification structure for your parts upon which you can build and scale with over time. All future acquisition data could then be classified and governed using this underlying taxonomy. From there, you can cleanse and enrich the parts data and load it into your PLM. This ensures that your team can quickly drive value out of the investment your organization has already put into the existing PLM solution.
The data is dirty. Many teams don’t focus on a data cleansing effort because they do not have the cycles to shift away from other critical priorities. The data fill and cleansing process can be extremely time-consuming, particularly for a team who doesn’t often focus on that task. Without the rigor of an established process, data cleansing efforts frequently fall short of expectations when stakeholders only tackle a partial sub-set of data or become distracted by other priorities. Over the years, CDS has developed the “secret sauce” to harvesting dirty data, normalizing it into a single standard, flagging issues, and loading it into a target PLM. This process evolved into what we now offer as an eight-step methodology to parts enrichment that begins with the ingestion of legacy data and ends with the newly cleansed and enriched data being loaded into a PLM. Many organizations have leveraged our methodology as an accelerator when their own cycles to cleanse and manage the data are limited.
The work doesn’t flow. Do you have automated workflows set up? Is it easy to locate where a new part is within your parts creation and approval chain? If not, there’s a far better way to track the creation of parts using workflows. Design for Retrieval (DFR), our leading solution, features a built-in workflow engine (and customizable workflows) designed to progress a newly introduced part along the NPI process and allows for reporting across statuses, from a higher-level category view down to an individual part. Manual tracking of activities, progress, and ownership can be cumbersome. Thankfully, we’ve solved for that with our SmartTrack solution, which helps to empower a data governance process across your parts data, driving user behaviors toward a parts-centric approach and ensuring your team doesn’t’ back into old patterns of working in offline siloes.
You can’t report on your parts. Do you know how many parts you have? Is there a way to quickly analyze what percentage of your parts have attributes? Many teams cannot answer these questions. For those that struggle with this, reporting on aspects such as fill rates, percentage of completion within an approval workflow, and ownership across parts within a workflow can make a world of difference in taking proactive and actionable next steps to parts data management.Analytics capabilities coupled with a centralized library of parts can drive a far deeper level of parts reuse and informed decision-making across your team.
Data from third party data sources is incomplete. While third party parts content providers may contribute some attributes to your parts data, that data is typically not complete. In short, the Engineering team will not have the full picture of a part’s attribution using that approach. By contrast, some parts may not be matched altogether with the third-party parts content, leaving your team with significant gaps that need to be reconciled (and that reconciliation takes time!). The other challenge with ingesting third-party data is the need to analyze the currency of a part. A given part may be newly introduced, in active status, have recently updated attribution, or may now be obsolete. When managing the sourcing and fulfillment of parts (and parts data), it’s important to know where a part resides within that lifecycle. This is where attribute clustering can also help. The ability to identify alternative parts based on the same or similar attributes can be a game changer for teams looking to optimize their sourcing and supply chain challenges. While third-party data is a starting point for many organizations, most teams will need additional support getting their parts data to be complete, consistent, correct and current.
Ready to take the next step on the journey to a more parts-centric existence? Contact us!
We'd love to chat more about how we can help and share some best practices with you.