June 5, 2009 - Blog Excerpt - Product Data Intelligence A New Perspective on Old Problems

"How long can I stay
In this nowhere café 'fore night turns into day
I wonder why I'm so frightened of dawn"

Bob Dylan from "This Dream of You" on his 2009 "Together Through Life" Album

These lines struck me on several fronts.  When I put my business hat on, I can't help but think of organizations that lament their current technology/solution challenges, yet are fearful of change.  Accessing part and related data and information is one of these challenges.

This "data stuff" resides in multiple databases and repositories and is composed of disparate structures, formats, notations . and can be as varied as the people who create and edit it.  How to find what consumers need?  Perhaps more importantly, how to harvest out of the richness of data critical to decision making knowledge?

Data warehousing initiatives seem to have checkered results.  Collect all the data needed.  Scrub and clean.  And then dump into a big bucket.  Gathering up the neighborhood cats for a bath sounds easier.

Enterprise application integration (EAI) efforts have also recorded mixed results.  Custom coding, couple by chasing revision updates and changing schema realities evolve and start to look like science projects.

Lamentations . groan.  Resources, maintenance, support and the ongoing costs of these two approaches would cause any organization to give pause to stepping out onto the field of exploring alternatives.

Let's face it.  Data is messy.  Applications change. Operational needs change.  Companies up-size, down-size, merge and divest.  People come and go ... and so does their knowledge of the products and processes.

So, what to do?  The paradigm shift exists in leveraging search based applications (SBA).  Acknowledge the diversity and varied quality of data.  Leave this data and supporting applications as-is.  Apply mature and proven search engine technology, parts-centric taxonomies, dynamic classification and parametric structures, and role based user interface tools.  The result, a product data intelligence (PDI) solution.

The challenges of manufacturing and engineering environments are unique.  Product data moves through lifecycles.  Various applications are used to create and edit this information along the way.  This includes such vertically focused applications such as CAD, product lifecycle management (PLM), electronic content management (ECM), and enterprise resources planning (ERP) solutions.

Each of these appliations have their own databases and schema, structure, processes and methods.  In addition to separate repositories, the nature and characteristics of the parts and/or products are further represented in such sources as requirements documents, CAD drawings, bills of materials, vendor and suppliers, test procedures, and so on.

The reality is that a single source of part or product data simply does not exist.  Internet focused search solutions have clearly demonstrated that consumers can find and retrieve information without any knowledge of where it exists.  Today this information can be text on a website, content in multiple document formats, and even derived from database driven sites.

PDI solutions take this concept further.  It builds on what users and consumers have grown accostomed to.  The most recent release of Microsoft's Bing illustrates the use of categories of potential groupings of the results.  It simply becomes another way to disect information.  Likewise, a PDI solution like Partrieve, takes a wide range of data sources and couples the results with dynamically groupings of this data.  As a result, users can quickly explore this combination of results and grouping to get to the items they are looking for. So, there's no reason to fear the "dawn". Get out of the "no where café" and jump into a low risk, low cost alternative.