PLM: How to Cut Tree Hierarchies and Empower Data Networks

by Oleg on August 14, 2012 · 2 comments

One of the most profound ideas in the history of PLM is the idea of PLM database that contains all information about product and its lifecycle. Major PLM vendors made a significant effort in establishment of centralized PLM repositories and product architecture backing up the idea and advantages of PLM database. In a nutshell, PLM DB represents the hierarchy of the data – requirements, product configuration, variety of Bill of Materials and not only.

Another idea that correlates to PLM database is the idea of the whole truth. I’ve been blogging about that in my post – PLM and the whole truth problem couple of weeks ago. Here is the conclusion I’ve made back in March.

PLM vendors need to learn more about last decade of web development and organization of large scalable web systems. In my view, an attempt to build a “singular” system won’t be successful and create a complex system that hard to maintain, change and scale. The future belongs to data networks and more flexible data organization.

The idea of PLM database is dominating now among PDM/PLM vendors. When it comes to discussion about how to put everything in a single data, you can hear voices explaining about integration and federation. However, in practice, it usually means an establishment of single mediated schema of data and centralized data repository. This repository is conceptually set of hierarchies. It comes to trees of product structures, variety of Bill of Materials, etc.

My hunch that one of the next PLM challenges will be movement from a single database power to the power of network. I conceptualized this idea during my AU 2011 talk last year – The new frontiers in PLM software. You can see slides here and video here.

Few days ago, my attention was caught by the following video: RSA Animate – The Power of Networks by Manuel Lima. Take your time and watch it. I found it quite interesting. Manuel Lima is a Senior UX Design Lead at Microsoft Bing and founder of VisualComplexity.com. More information about Manuel’s work related to the power of networks is here.

The video brings some very powerful examples related to the shift from hierarchical organization to network organization of information, data abstraction and data. I think PLM data architects need to think how such type of information organization can help to PLM implementations.

What is my conclusion? The complexity of data in product development and manufacturing is growing. In order to solve the problem of data complexity, PLM vendors need to find new models that can scale up and leapfrog the current limitations of databases. Network model is much more powerful and promising in this context. Important, in my view. Just my thoughts…

Oleg

Image: FreeDigitalPhotos.net

Share
  • http://twitter.com/avwikinomics Michael Wm. Denis

    “The complexity of data in “product development and manufacturing” is growing.” So when PLM vendors move more into the Service Lifecycle Management (SLM) arena – such as PTC's acquisition of Servigistics – this complexity skyrockets exponentially.  SLM configuration trees are much deeper / more granular than PDM/PLM hierarchies. This is a requirement of component level reliability centered maintenance (RCM) task scheduling and execution – includes component level performance data. Additionally, component level task & sub-task content is managed in highly structured XML schemas / native dB. Synchronizing SLM dBs with PLM dBs given the differences in data types and structures (due to differences in data usage and intent) means any “holy grail” of a singular repository will never happen for Total Lifecycle Management (PLM + SLM). What does need to happen is the ability to synchronize changes in configuration management.

  • beyondplm

    Michael, thanks for your comment! Absolutely agree with you on the complexity between PLM and SLM. The same happens when you synchronize changes between PLM and ERP environment. Love your comment about “the holy grail” of a singular repository. In my view, there is no other way than “share and link information”. Two previous options – 1/synchronize; 2/single repository failed. Just my thoughts… Best, Oleg

Previous post:

Next post: