What is the Role of Search in Enterprise Systems and PLM?

January 30, 2009

Going back to the early days of enterprise software, vendors developed all  they needed from very basic operational systems features. This is was the only way to implement functions of the system. Very core functionality of the system was data storage – the heart of each enterprise system. Storage was proprietary and optimized for the type and structure that data system were expected to manage. The real revolution began with the movement to database management systems (RDBMS). These provided a standard way of managing data and accessing it with SQL (Structured Query Language). For the last 20 years,  I see that this situation has remained unchanged. Enterprise software has created unpredictable huge data stacks managed by complex models and systems. During the last few years, active development of SOA related technologies have also focused on how to make enterprise software more manageable and lean.

With all the complexity of enterprise systems, I don’t see much involvement of search-related technologies within the enterprise. On the other hand, the search revolution has changed consumer internet, market, and actually, our lives (Google Map and beyond). I think that the future of search in enterprise is still unfolding. Enterprise systems are real gold-mines for search systems today. Search is really underperforming and focusing largely on the discovery of various documents  and document content.

At the same time, I’ve seen some movements within enterprise search, and search in general. Multiple “verticalization” of search is focusing on how to provide vertical domain specific searches. Examples are searches for healthcare, consumer market niches and others. “Horizontal” activities are investing into semantic searches as well as information discovery areas. A separate focus is the delivery of search appliances (of course, with a heavy investment of Google in GSA).

I think that searches need to leverage technology and capabilities developed for Web and unfold enterprise data for users. PLM might be a good starting point. Today’s PLM implementation survives from the need to consolidate different domains of activity around product development – starting from product requirements and customers, and ending  in product recycling and regulation. Search can become a good platform to make this happen. The complexity of enterprise systems and their openness is a real challenge for search technology, and, as typical in this case, leaves a  lot of space for innovation.

CAD is boring… because CAD is alone!

January 29, 2009

I was reading an interview with Chris Williams about his new product, Vuuch, on NOVEDGE. Chris said that “the CAD world is boring…”. But I’m wondering why? I thought that the data management world (PDM, ERP) is more boring since it’s all about grids of data.  But CAD? This is about really great pictures of nice cars, executive jets, incredible yachts, etc… then I started to realize what happened. I think CAD developed as a very deep silo, from the user standpoint as well as from the organizational standpoint. CAD does not allow people to expose everything that CAD creates to the outside i.e. the engineering world. As you know, almost everything from the design world ends up in PDF format. Because CAD doesn’t exist in the outside world, a lot of intelligence, information, design content just dead after result of CAD goes to outside world.

I think that association with MRP/MRP II systems is very appropriated in this context. In the beginning of MRP, the scope of MRP was to manage manufacturing materials. Beyond MRPII, was the idea of managing a wider scope of data – materials, finance and resources. The shift from MRP to MRPII allowed the connection of these three data silos together. Future transformation from MRPII to ERP added additional scopes of data and processes beyond manufacturing to be integrated all together.

So, what I see as the fundamental goal is to allow CAD content to be available in various forms outside of the design space. CAD silos need to be connected to other aspects of engineering and non-engineering activities. This will build more business value around CAD and ultimately create more value for organizations and users. I think this space is open for disruptive innovative technologies that will make changes in this space similar to introduction of parametric modeling in CAD, creating of 3D CAD on top of Windows, and some others. Today’s design content requires modernization, using more efficient and cheap technologies that will link the outside world to CAD.

Clayton M. Christensen, author of “The Innovator’s Dilemma” shows that sustained innovation, even funded by billion dollar research budgets, only enables leading companies to survive as long as no disruptive technology suddenly appears to upset the corporate apple-cart. I believe that new PLM, Digital Prototyping, BIM or any other technologies that will come to this space need to be more social, more open and more reliable compared to what we have today. And… CAD 2.0 or PLM 3.0, or whatever TLA (three letter acronym) product will be developed in today’s world, in my opinion, will definitely not be boring…

Check out Ivan Sutherland’s movie on Sketchpad. How do you feel when you look at your 15 year old computer today? Funny, isn’t it? I’m sure that CAD/PLM of today will appear just as funny in the future as the CAD of 25 years ago seems today.

Use Predictive Modeling to Decrease Product Cost during Product Design and Development?

January 28, 2009

We are all concerned about how to decrease product cost. This is a top priority of users in today’s economic situation. But it was also a priority before 2008/9. Our primary goal is to provide systems that allows control over the cost of products. Since 80% of product cost is already defined at the stage of design and early product development, predicting cost, in my opinion, should be on the short list of PLM product developers and implementers.

 “Predictive modeling is the process by which a model is created or chosen to try to best predict the probability of an outcome” (Wikipedia).  Sounds complex, right? But predictive modeling technologies based on statistical data analyses are widely used today. Many systems analyze historical data and predict future behavior. A similar example of predictive modeling usage in PLM systems is evident in Customer Relation Management (CRM) Systems are analyzing customer level models to predict customer behavior in the future. For example, health care systems analyze existing customers to predict high-risk members; telecom operators use predictive analyses for cross-sell opportunities (by analyzing product combination patterns purchased by other customers), as well as customer churn. 

Now, how do you implement this? In my view, this is all about connectivity between your systems. Today’s design and engineering are very localized and have a limited view on what is going on outside, how customers uses their products, what happens with service departments, etc. If systems are to establish a connection between particular design decision and customer defect reports, extra expenses by suppliers etc., this information can be potentially reused to predict future cost and product defects. Even if it sounds ahead of its time, I’m pretty sure our future is there. More practical examples of predictive engineering are in the areas of FEA and other engineering analyzes. Accumulating statistical data in this area can provide good prediction results for product design connected to customer experience.

 I’d be glad to discuss with you potential scenarios of predictive modeling usage and… may be you already have some of them implemented… who knows?


Contextual Collaboration within the Organization

January 26, 2009

I’m thinking about how to improve communication within an organization and across the professional eco-system. For example, let’s say you are an engineer in an average-sized organization of 1000 to 2000 people and you work with a few dozen suppliers. Let’s assume that you have a particular design issue and would like the advice of relevant people, and want to involve them in a discussion.

Now, picture the situation: You write your query or request, attach or associate all the relevant documents, and then press…. <Send>. What is missing from this sequence? Yes, the list of people with whom you want to collaborate. In a typical organization, sometimes it is not that simple to identify relevant peers. Also, your memory might not be as good as you want it to be in this situation, especially when your professional eco-system is large. Also, don’t forget that our lives are very dynamic. So, here is my point – a system can identify to whom this message needs to be delivered.

Sounds good?

Then let’s think how we can make this implementation happen. There are a few available technologies that can help us:

1.     Natural Language Processing.

NLP, or a special sub-set of NLP called Information Extraction, can be used to retrieve key information from the message that you composed that will help the system create an initial set of criteria for searching relevant parties in order to deliver the message.

2.     Annotation

The broad meaning of annotation is adding of information to resources. There are different ways of annotation in real life and over the Web. For example, you can annotate Web pages to help the search crawlers index them according to particular needs and/or meaning. In our case, we can annotate people (contacts) in our corporate system or global addresses / contact books. This information can be generated based on analyzing corporate organizational information as well as by the contribution of the people themselves.

3.     Search Engines.

You need some way of indexing all possible annotations to make them available and accessible. Afterwards, based on key information retrieved from the message, you will be able to get all relevant people to deliver the original message. Depending on how deep you want to go into indexing and information dependencies, you can use multiple options here. If you are in an organization that has Microsoft Office SharePoint Server, you can try leveraging the Enterprise Search from MOSS or any other Microsoft Search product. If you use Google Fun, you can think about Google custom searches or try Yahoo BOSS for Yahoo.

If you just want to match strings, you can put this information into a relational database and it will work for you. If you want to be purely simple – use MS Excel services :)… of course, some people tell me that everything in this world can be done with Excel.

Actually, I didn’t plan to implement this system as part of this blog post, but only figure out some core issues about how to make collaboration smarter and save a few minutes in searching for the relevant people with whom to communicate…  So, if you like this idea, let me know – maybe some of you already have something similar in place within your organization? 

How to improve PLM collaboration and information delivery with RSS

January 23, 2009

The issue of information overload is growing in importance. People are getting more and more information via different sources – RSS being one of them. You probably use it to get feeds from blogs, news and other internet sources. Actually, RSS technology is very scalable and, in my view, can be used by PLM systems to delivery information to the right people. Forrester researched the space of RSS usage inside enterprises – indeed RSS can aggregate content and distribute it to PLM system users. What I like in this type of solution is that information can be delivered to internal and external (value chain) users in the same way, using an RSS feed.



 Conclusion: In the world of Web 2.0 and PLM 2.0, the usage of RSS feeds to distribute information can solve problems of data flow and information delivery. RSS is still not adopted by all users. What I found surprising is that there are also not that many software vendors providing an RSS solution for enterprises. An example of an existing solution is the company called NewsGator (http://www.newsgator.com/). This company- provides a solution to boost collaboration and improve social communities in your enterprise. So, the solution is relatively simple and available – you can try to use it with your PLM system. I expect that there may be some existing PLM providers who can support RSS as part of the standard collaboration functionality.


PLM in economic downturn – Is there a place for second-mover innovation?

January 22, 2009

How PLM can continue growth in today’s downturn economy? I think, this is time to get back and talk about PLM user adoption. I’d like to raise this question in a very specific way – is there an opportunity for second-mover innovation in PLM today? In today’s market situation, second-mover innovation can open back-core PLM areas and allow the delivery of PLM systems (or components), leveraging all research done previously by the first movers of PLM innovation, who are the majority of PLM vendors.

 Second-mover advantage occurs when a firm who follows the lead of the first-mover is actually able to capture greater market share, despite having entered late. First-mover firms often face high research and development costs and the marketing costs necessary to educate the public about a new type of product. A second-mover firm can learn from the experiences of the first mover firm and may not face such high research and development costs if they are able create their own similar product using existing technology. A second-mover firm also does not face the marketing task of having to educate the public about the new project because the first mover has already done so. As a result, the second-mover can use its resources to focus on making a superior product or out-marketing the first mover…the notion that winners are always the first to enter the market is a misconception. (Wikipedia entry)

Examples include Obama (vs. Hillary), Nintendo (vs. Atari), Oprah (vs. Donahue), and AMD (vs. Intel). Second-movers are able to capitalize on the work done by others and create innovative products at a lower cost. The “early bird may catch the worm” but later birds can dominate marketing and distribution.

Assuming that we are ready to research existing PLM spaces for second mover innovation, what should we focus on as the ultimate functions/components of a PLM system? Presently, I see, Bill Of Material Management as the top candidate for second movers. Most of PLM starts and ends with the management of Bill of Materials. The ability to manage multiple Bill of Materials is a requirement of all customers interesting in PLM-like deployment. 

If the idea is to re-invent the Bill Of Material portion of PLM, what should be improved, in my opinion?:

 Bill Of Material definition: The ability to craft any Bill of Material without a preliminary definition. The Data Model for BOM needs to be open. 

User experience: Flexible rule based user interface, similar to Excel, plus the ability for customization

Multiple Bill of Material synchronization and rules: The ability to manage and synchronize more than one BOM.

Data acquisition: Multiple ways to acquire BOM data – from CAD systems, from other PDM/PLM tools, ERP etc. Practically, BOM can be created from any place.

Bill of Material module encapsulation: To invest into SOA and interoperability of BOM implementation with the rest of the systems in PDM/PLM, and the extended domain, ability to use BOM module with various collaboration systems such as SharePoint, Lotus family etc. 

By doing that, BOM will be converted from existing rigid definition and implementation to modern tools, capable of handling Bill of Materials in various scenarios and user cases. 

I believe implementing such a “back to roots” strategy will allow the creation of significant market advantages for increased user adoption of PLM. Second-mover innovation is relatively cheap and can leverage all market investment done by PLM companies to date. 

Who will do it? This is an open call… 


Get every new post delivered to your Inbox.

Join 290 other followers