I’m continuing with a set of ‘back to basics’ questions and discussion. My topic today is about singularity or what known in PLM as “a single point of truth”. I remember the time before computers were widely spread in engineering and manufacturing organizations and people used “Drawing on the wall”. The famous statement back to that time was – if you see drawing on the wall, think maybe there is “another wall”. Computers supposed to resolve the problem of multiple walls. However, the problem becomes even worst. Now you may have an unlimited number of “electronic walls” in your email system, disk on your laptop, server in your location or different enterprise systems. PLM comes with a very promising and illusive statement about single point of truth. From PLM standpoint, it defined as capability (or ability) to manage single (or unique) version of product data in your organization.
Now, my questions today about how to manage this uniqueness? Is it something you can see as an achievable goal for your next PLM implementation or wishful thinking? Is it something we can realistically implement in the organization with all systems we have in place? Can we achieve it in a single shot, or we need to have multiple steps to get there?
Here is my take on this (and I’m making it the most provoke as I can think about).
1. PLM Singularity (or single point of truth) represents a fundamental interest of people in the organization to organize and manage product data.
2. We can implement it in the organization, but it requires a significant effort that may vary depends on organization complexity and number of other systems involved into product development process and number of geographical locations company plan to cover by this system.
3. Implementing of singularity may take multiple steps until it will be achieved in the organization.
So, what is my conclusion today? I think, singularity cannot be considered as a goal for PLM implementation. The final cost can be too heavy to carry, and it will take long time to get there. Realistic PLM implementation need be done as “go by steps” approach. Each step will achieve goals helping organization to resolve specific issues they have in the organization or optimize existing processes. This is the only way to create something I’d call – lean PLM. If you want to implement single point of truth, think about playing “singularity chess” before :). (thanks, Paul Burgess blog for picture)
Just my thoughts. I’m looking forward to your comments.
Best, Oleg
