3 things PLM can learn from UCO (Used Car Ontology)

December 18, 2012

Our digital and real lives are getting more and more connected. Think about our web experience these days. Mobile devices, map service, location based query, social networks. The list of examples can continue. One of the biggest challenges we have, as a result of web exposure in our real life, is the need to integrate and interconnect many sources of information coming from different places. Think about intertwining of your location information, photo posts and reviews. You have Facebook Nearby service (the enhancement of this service was just announced yesterday), You can get some interesting perspective on the service here.

Moving to the main topic I wanted to discuss today, navigate your browser to the following SemanticWeb article – Introducing the used car ontology (UCO). The article speaks about publishing of ontology (knowledge of concepts) that supports a precise description of used cars. The article reference MakoLab, the company that made this work as well as provides a link to the ontology itself. I found the following passage interesting.

“Publishing information about used cars… containing a description which refers to the Used Cars Ontology will allow for easier searching of cars for purchase, along with a more in-depth description of their state of exploitation. The UCO supplements the more general ontologies GoodRelations and Vehicle Sales Ontology created by Professor Martin Hepp. The GoodRelations ontology is now integrated with the famous dictionary schema.org created by Google, Yahoo and Bing, for the purpose of improvement of information searching on the Internet and better positioning of websites.”

Dig a bit into UCO ontology document and you find some examples of queries and operations build with the use of UCO ontology. Spent some time with the document and you will learn how to get report of used cars, getting car description, updating information about the car and more. The information about used cars can be located on multiple websites. It made me think about possibility to improve interaction with multiple island of information. Here are top 3 things I came after the analyzes of UCO doc:

1. Ontology can be used to produce a meaningful queries. The web technologies are providing a reliable instrument to work with data located in multiple websites. Semantic web provides set of technologies helping you to describe, query and manipulate the data.

2. Nobody is interested how data is stored. It is almost meaningless how data is stored. The website with the information about used cars can use any technology (from text files to excels and databases) to store data. This information is not needed to process data on a web scale.

3. Publishing semantic information can improve cross system data access. When your website and/or service is publishing information in semantically accessible way the information can be intertwined and used by other services for different purposes.

What is my conclusion? Web is a good example of the system that grew up beyond the level of single database. Web data processing mechanisms are interesting from the standpoint of sustainability and data scalability. Used car ontology provide a good example of organizing interoperability beyond the level of a single website. My hunch, we are going to see some of these technologies can change the way PLM systems operate today. Just my thoughts… What is your take?

Best, Oleg

PLM IP Management and Tribal Knowledge

June 20, 2012

IP Management. If you’re in PLM business most probably you had a chance to hear this term. Most of the PLM providers are using "IP Management" term to explain the way product information and related data in the organization are captured and processed by PLM software. You can see some examples from Aras, Wipro, Dassault (by CIMdata), Kalipso, PTC, etc. Google PLM IP Management to find more. I found the following passage by PTC about safeguarding IPquite explaining well the situation related to IP capture:

Today, IP contained in product concept designs is increasingly at risk, along with the revenue and corporate know-how it represents. The threat to product IP has increased as a result of globally distributed design teams that design products collaboratively, and ironically, by the approaches required to enable truly distributed design and manufacturing networks. A lack of processes and procedures in place to protect IP and safeguard design concepts during development, can lead to lost IP, which in turn results in lost sales, product commoditization, and lower profit margins.

Design is clearly one of the sources of IP. CAD, PDM and PLM systems today can help in capturing of this information. We often use term "structured data" when talking about sources of information like 3D Models, Drawings, Bill of Materials, Manufacturing procedures. At the same time, it made me think about the fact lots of IP information lives outside of so called "structured data". These days the topic of capturing IP from other sources of information becomes very important.

I had a chance to read semanticweb.com article From Business As Usual to Knowledge-Driven Architecture by Yefim (Jeff) Zhuk bringing some valuable insight on processes of knowledge capture in an organization. Interesting enough, structured knowledge was estimated as 10% of overall knowledge in an organization. Another 20% defined as "unstructured". The larger portion of knowledge (70-80%) defined as something called "tribal knowledge". Here is the definition from the article:

Structured data are very formal. They are defined in terms so precise that not only people but even computers can understand these data. Databases, Business and Data Models, Services and XML Files are structured and understood by computers. Unstructured data are documents and communications artifacts, like taped messages and video clips that make sense to people. The knowledge captured in unstructured data is hardly available to computer systems. The biggest portion of information that is used daily in business routine has never been captured. It is so-called “Tribal Knowledge”.

The sources of "tribal knowledge" are related to communication between people, meetings, phone calls, etc. These gapes are increasing significantly when we are moving from a single (larger organization) to the network of smaller organizations (SME). Go to every small manufacturing firm and you discover that amount of information about development processes and decisions made never been captured.

What is my conclusion? The awareness of the ability to capture information and future re-use is growing. People understood the value of information these days. Big Data is just one of the most trending topics today and related to the ability of computer systems to capture information online related to social communication and other web and organizational data. "Tribal knowledge" is one of the them. In my view, the intelligence of manufacturing companies is related to the ability to capturing more information and future decision optimization. Just my thoughts…

Best, Oleg

Image(s): FreeDigitalPhotos.net


Get every new post delivered to your Inbox.

Join 290 other followers