I wanted to get back to micro-blogs, and more specifically, to Twitter. In one of my previous posts, I discussed the potential of micro-blogging as a tool for collaboration. I’d like to take it even further, using Twitter as an example.
My observation is that Twitter can be a good example of how to create a collaboration tool for people who dislike managing their work, by definition. I’m referring to designers and engineers…
Why do I think that Twitter, in particular, and micro-blogging, in general, will make designers and engineers happy? I think that the main point is that tools like these will free their mind while keeping a record of their work at the same time. A Twitter-like collaboration tool will allow them to communicate openly. However, on the other hand, Twitter provides very good notification and search capabilities (take a look at http://search.twitter.comor http://www.tweetfeed.comfor example).
So, here is my basic proposal for Twitter as a collaboration tool for designers/engineers:
1.CAD/PLM content can be placed or referenced in twitts (by reference, 3D or picture)
2.Allow designers and engineers to manage conversations (following, direct messages, notifications).
In today’s world, more and more companies need to collaborate globally. In the supply chain, this is a result of companies wanting to optimize their operations and having more suppliers involved in all processes of their activities. There are many solutions that companies are using for the global supply chain. With the current development of online tools, internet, and services, most of these tools have established online spaces where you can share and exchange information and transmit messages.
But something makes me feel unhappy about these tools and spaces. What I dislike is that most of them create their own eco-system. I think there will be a significant improvement if we are able to connect these solutions to organizational processes. But this is not always possible because companies today do not share their processes. In most of the cases, they use different tools to implement business process management in their organization.
How can you make this collaboration possible? In my opinion, development of process interoperability can provide a new dimension in how company processes work together. The ultimate requirement is to keep processes separate, but, at the same time, allow them to work together. The main problem of such integration, as I mentioned above, is that because business processes are separate, data-base schemas are different. Each company process uses its own data schema to represent information which is tightly bound to a specific process implementation.
Today, I can observe two major ways process integration occurs: (1) mapping based approach; (2) query based approach.
For the mapping based approach, the integration creates a map between two local schemas. By doing this, a new “federated schema” is created. In some cases, companies are trying to use one of the schemas as a federated one. This is not always possible as sub-sets of schemas are used. But whatever the final schema is, this process is complicated and requires multiple mapping operations to happen. This way is also expensive, in my view, for long term support and changes. You can see some examples of this approach here: http://www.research.ibm.com/journal/rd/485/zhu.html
Another approach is query based. In this approach, a special interoperable query system needs to be developed. This “query” is able to establish a connection with both (or multiple systems) and extract/load information. With the latest development of SOA and Web Services, this way has become very popular. You can see example of this approach here: http://www.oracle.com/applications/pip-for-oracle-product-hub-data-sheet.pdf
I’m thinking about different way – integrated ontology development. This ontology would place a super-set on top of both business process models. Ontology, in simple words, is a data model. So, we need technology and/or a tool that wold allow us to develop an intermediate data model. There are such technologies in today’s world developed as part of Semantic Web W3C initiative – technology based on RDF (Resource Description framework) and OWL (Web Ontology Language).
These technologies have been developed for the past 7-10 years and, in my view, have become more and more popular. They allow you to establish a global model for content and message transfer, thereby simplifying communication. The main advantage of this collaboration is that a new level of data models will be created. These data models, represented by RDF/OWL, will allow companies to establish a connected data environment that can be used by multiple systems.
And, to make some fun, take a look on this video. Even if content is not related to supply chain yet, the core idea of integration is the same.
Today, I’d like to discuss the relationships between Product Lifecycle Management and Enterprise Collaboration. Enterprise collaboration continues to be important. In today’s world, the ability to collaborate is critical for all manufacturing companies. Top enterprise collaborative tools include email, instant messaging, audio conferencing, web conferencing and video conferencing (see The State Of Real-Time Collaboration In 2009: Conferencing Tools Are Hot). We can see that all these tools are focusing on communication between different departments in the organization as well as between other organizations. Organizations of different sizes may prioritize their collaborative needs very differently – for larger companies, enterprise collaboration together with enterprise search is important; for smaller companies, consolidation of existing tools and email systems is a top priority.
So, how is PLM connected to the enterprise collaboration story? Should PLM develop collaboration on its own? How can we consolidate efforts and what can we do in the short-term? I think that enterprise collaboration tools today are not completely ready to be used together with design, engineering and manufacturing systems. Their common uses, in most cases, are limited to very loosely coupled mail communication, and these require a great deal of effort to make them work together efficiently. At the same time, there are quite a few PLM collaborative tools that are available and allow you to work together on different tasks such as co-design and co-review, thus letting you communicate among people.
I see that PLM user adoption can be significantly improved by integrating PLM tools into various available collaboration tools – mail, instant messaging, video and web conferencing, enterprise search and others… Now, what is are the key points for connecting all of these? In my view, there are three major points that need to be implemented in order to leapfrog into PLM collaboration. All three points revolve around content. The priority of content is very clear in my view.
So, what can PLM do?
1.Improve content availability
Content needs to be available at the same level you use a word processor today. Imagine simple copy/paste functions available across all collaborative tools.
2.Allow contextual collaboration
When content becomes available, the next step is to allow people to communicate on the same piece of content in the context of particular task, operation, activity. This will simplify task management for many things people are doing together.
3.Keep a history of collaboration
Last, but not least. We need to keep track of what we are doing. How? Let’s keep a history of our operation, actions, and changes. This can improve significantly what was done, when and why.
I think, by doing these three steps, we can significantly improve PLM tools and implementation.
There are several examples: Multiple viewing and reviewing solutions – DS 3DLive, JT Open, PTC Product View, Oracle AutoVue (former Cimmetry), Actify and others allows collaboratively design, review and work on 3D information. Still, in my view, work needs to be done in usability, granularity and search-ability. The point that I feel is missing is the lack of asynchronous collaboration.
Another example is multiple publishing tools: 3DVIa composer, Right Hemisphere, Adobe. These tools allows us to transform content to different formats and make this content widely available for additional users in the manufacturing eco-system. Still, in my view, lots of improvement in the areas of data unification and reuse still needs to be done. Most of the formats are incompatible. Transformation between these tools requires a large deal of implementation and is not always available. In most cases, publishing formats do not allow you to make changes. If they do let you, these changes can barely be synchronized back to design tools.
What is your view? How do you implement these tools into your organization?
As you know, new ideas often come from things already forgotten… Therefore, I like looking at research projects and experimental products as they are good ways to find some fresh ideas. A few days ago, I was looking at the Lifestream application created by Gelernter 1994; Freeman 1997.
Another predecessor and association is Malone’s influential study – “How Do People Organize Their Desks?” (1983). I found some very interesting concepts that, in my opinion, can be valuable for Product Lifecycle Management. What particularly caught my attention was the association between lifecycle and Lifestream. Actually, in regards to lifecycles, I think that we are very disconnected from representation of time in PLM applications.
Lifestream is about time. Originally, the proposed approach was to throw away filenames, folders and static filling and replace it with stream of information arranged according to time. Actually after the implementation of Lifestream, some of its features were found in Apple iApps and other commercial products.
While terms such as file/folders/desktop are brilliant, but Lifestream focused on the non-traditional aspects of information management, which is not typical for regular computer users. On the same point, I think PLM is different as a good concept of managing of product data that needs to be invented.
The Lifestream model is about to create, organize and present timelines. Whereas timelines may have different representations – collections, calendars or queries, this model would eliminate all other artificial models that people need to create in order to store product information. It’s interesting that this model would allow you to use a “time oriented” memory. An additional benefit of this approach is that the timescale lets you use the same model to manage tasks by leveraging future timelines. Appropriated information “tagged with future date/time” would appear in the scope of the system whereby the user would get this information as an alert.
I think this is definitely an interesting concept that requires additional research and validation with users. I’d be glad to hear your comments and thoughts.
One of the biggest PLM challenges is to play the role of connecting “organizational dots” such as Design to Manufacturing, Engineering To Order, Configuration To Order, Concept to Manufacturing. All of these terms are about creating a world where designers and engineers are connected with each other as well as with other players of product development – manufacturing, suppliers, and consumers. However, this connection is not straightforward – design and engineering data sometimes seem like a foreign language.To some departments, communicating with another department is like tourists trying to speak a language they don’t even understand.
The biggest problem I see today is that CAD and engineering knowledge remains in the Engineering Department language. A very small amount of people can speak this engineering language, not to mention knowing how to connect an engineering solution to solve a business problem.
At the same time, everybody knows about this problem. As one of the examples, take a look at this Adobe blog which presents the challenge of enterprise data sharing.
Out of the many different solutions that try to resolve this problem, I’d like to touch on a few them which I think are quite successful:
PDF and 3D PDF based tools are definitely leveraging the high PDF adoption rate by translating CAD and related engineering data to 3D PDF packages. You can use various offline and batch processing tools.
In parallel, XVL technology from Lattice also provides an alternative to translate CAD data into an XVL format and share this information across the organization.
3DVIA composer is also a translating tool – it allows you to gather CAD and 3D data together with additional information related to technical publications, manufacturing and some others. After generating new 3DVIA composer files, you can share and reuse this information in the organization.
An alternative approach in DS is the 3DLive product that allows everybody to access 3D information online without translating this information into other formats.
I can bring additional references on the tools that take 3D and engineering data and allow reuse of this data into enterprise. But this is not the key. For the past 20 years we succeeded in creating lots of tools that translate one data into another data – CAD to PDF, one CAD to another CAD (i.e. IGES, STEP), etc. My key point is that we are still missing the capability of connecting people working on different problems in the same organization: people working on design, people working on various manufacturing and supply chain optimizations, sales and others. There is no technology that allows them to work together, share the same problems and see one solution. We have come a long way – from proprietary data storage, RDBMS, XML, various 3D proprietary formats, various “standard” formats etc. All this being said, we are still missing something that can be used as a scalable foundation to represent CAD and engineering knowledge.
This technology hasstill not been invented, in my opinion, and this is a big topic. Since I’m sure I’ve missed references, examples and ideas. I’m planning to continue posts on this topic. I’d be glad to hear your voices and comments.
We live in a physical world. Everything in this world seems to be created in a parallel, virtual space. This space is in our heads, in the computer systems we use to design this physical world, and in the global enterprise systems that manufacture, support, and dispose things in this physical world. Certainly, there is a connection between these worlds that we can discover and use. But how? I’m sure this is a long journey. In this post, I will think out loud about some possible technologies that are available or emerging today to help us to establish this connection.
Practically, to create links we need to establish two elements of connection – (1) stable identification system; (2) technology for keeping the identification of this information.
Identification system
Probably the best resource identification system today is the URI based system from which the Internet was founded. There is some interesting research that I found in the area of creation of persistent unified resource locators (PURL). Such a system can be potentially adopted to indentify virtual objects. This system eventually can be used to enhance identification of today’s virtual models created with CAD/PDM/PLM and other manufacturing systems. On the physical side, technologies like RFID cards, QR (Quick Response) codes and smartcards can be the most common.
Proxy objects
These identification technologies can be connected together by proxy objects. Information on the virtual side can be organized in the various cloud services and locations as an organized network of virtual objects describing everything that has been designed, manufactured and has existed. Now, on the physical side, various technologies (cameras, RFID readers etc) can be used to scan information from physical devices. What is missing is a sufficient amount of cloud services that will allow us to store and keep track of this information online.
This idea is a little futuristic, but considering some of the products already available, I can see that some already show ideas and prototypes for this futuristic future. For example, an idea is the ability of created physical objects to react to their environment, – is something I may be expand upon in future posts.
Below are some examples of existing development and applications that inspired me in writing this post.
(Bruce Sterling, SIGGRAPH 2004, August 2004.) Bruce Sterling’s speech at SIGGRAPH 2004 describes a vision of objects that are connected to information related to their design, creation, and use; end-user reviews, ideas, and improvements; and where they are at all times.
(Kristi Heim, The Seattle Times, 31 March 2008.) Researchers at the University of Washington are exploring the positive and negative aspects of using RFID tags to track the movements of people in a social setting — by tracking themselves.
Researchers at the University of Florida are developing an ambient system to measure a person’s vital signs as he or she enters the house, transmitting the information to family or doctors, as a way of monitoring at-risk individuals or the elderly.
And one more… The book (Shaping Things) by Bruce Sterling influenced me tothink differently about the world in which we design, manufacture and live today. While Bruce is indeed a futuristic writer, a lot of good ideas in this world originated in futuristic writing. I hope you will enjoy the following interview with Bruce about his concept of SPIMES- virtual and physical objects:
In my view, open source (or more precisely –open source software) is becoming more and more visible in the business of enterprise software.
There is an important characteristic of PLM related to business that we need to keep in mind in order to analyze the impact of open source on PLM. There is a high segmentation of customer requirements for different industries and customers needs for different types of customers. Today’s “mindset PLM leaders” still are not capable of covering the full scope of customers’ needs. This topic makes open source both attractive and disruptive for PLM.
On one hand, when there are many common customer requirements, open source is very beneficial. Users have an ability to submit code from all participants of open source, forexample, as in a Linux system. This scale up is very significant and allows us to develop big systems according to communities of users. But, at the same time, if there are quite many different requirements this can disadvantage of open source.
In addition, due to the high percentage of services and customization, the ability to deliver overall PLM project as services without licenses cost is a significant advantage and changes the game of PLM system implementation. So for this type of customers, open source will probably be beneficial. At the same time, smaller customers will be more interested in standardized functionality with fixed cost that will allow them to achieve their ROI goals and less interesting when implementation turns to services projects.
My conclusion – on one hand, open source in general and open source PLM specifically brings a significant improvement in the way we create systems for our customers. These models shift the initial discussion from license selling to customer requirements. On the other hand, it’s not clear how IP protection in different organizations will allow the crowdsourcing of open sources and common delivery in this field. This happened successfullywith different non-PLM software, so it’s definitely a place we need to watch in the future.
I’ve been observing a significant increase in the availability of personal services in today’s software and solution offerings. Many examples of social design are focusing on the ability to design products with a “personal touch”. In addition, there is a very strong trend called “personal web”, which is collects content, tools and capabilities from the Web that has configured and focused on a particular user. It started from personal pages and configurations and continued to provide tools for tags, categorizing, sorting, searching and re-assembling (aggregation). People create a customized environment and tools are focused on the same process.
Now, how can this be related to Product Lifecycle Management? In the past, we considered PLM as an enterprise system that focused on business processes and followed product kifecycles from the earliest idea time through to disposal. Is there a place to inject “personal PLM” into current concepts of PLM? Does it make sense for PLM to put spotlight personal activity rather than the overall process? Don’t get me wrong J… I’m not saying that processes are bad in general. I’m just saying that processes need to happen, but people need to focus on their personal activities and the content that they need to create.
So, what is the idea? I was looking at some of the today’s examples of personal web in the areas of community building, self learning, and library management. I find this example of PLE (Personal Learning Environment) as interesting as the prototype of a personal PLM system.
Another example of personalization is the creation of widgets and tools that allow you to aggregate PLM content differently. Take a look on the following example of pageflakes (a system that lets you aggregate content from different Websites).
Product Lifecycle Management Tools today (even those tools that aren’t referred to as PLM) provide a very traditional set of tools focused on design (i.e. CAD), engineering (i.e. BOM), simulation (i.e. CAE) and other domains. If we can find a way to re-aggregate capabilities provided by these tools using modern SOA technologies to create a personalized combination of tools or a working environment, we can significantly improve the way people work today in product development
I’ve been thinking about how a company can use business process tools to manage their processes. In one of my previous posts, “How to improve PLM process before PLM system using BPMN,” I discussed the possibility of using process tools supporting BPMN to capture business processes in the organization. One of the key ideas in this post was that you can use BPMN tools to capture process definition, and later, to implement these processes using tools and systems you already have in your organization. PLM and dedicated BPMS products are systems that can be used for this process implementation.
When I started to think about how process capturing may occur, I came to the conclusion that the business process capturing process is very painful for an organization. As for product development in a manufacturing organization, this process can differ greatly from organization to organization. So process capturing is natural step for the organization when they implement Product Lifecycle Management and create a collaborative business process environment.
What we can do to change this? One of the ways is to reuse the practice of social networking and crowdsourcing for process capturing. My key point in formulating this conclusion is that process development knowledge in an organization is spread over many departments – engineering, manufacturing and others. It is very problematic to have people efficiently involved in capturing their existing processes – too many people, too much time… Also, people are busy running their businesses and claim not to have time for processes capturing. Allowing all people in the organization to be involved in ‘capturing processes’ which I refer to as “Process Thinking” changes the rules of the game. With everyone involved, people are able to do multiple reviews. Step-by-step adjustments can make a significant change in the product development process implementation for PLM. In addition, this process can improve people’s process adoption. Socially created processes will reflect the existing processes accurately. Afterwards business process management tools can optimize and improve existing processes.
How can we make this happen practically? The answer probably comes from software providers. What if process tools were to support a staged process of continuous process changes? Each person would be able to adjust the process to reflect the way people manage their work. Allowing everyone to vote and approve process changes will let all people be involved in the process definition possible right from the beginning. Another important achievement would be an increased trust in the process definition – normally, people trust something more if they were a part of it and defined it collaboratively together.
I believe that you may have some experience in this by trying to centralize the definition and approval of processes. However, making process definition open and transparent for all people in the organization can allow PLM companies to leapfrog the overall adoption of PLM systems in an organization.
The development of game technologies is taking huge step forwards and is sometimes outperforming product development tools such as CAD, simulation etc. It is very interesting to see how game software has new capabilities that allows you to visualize, simulate and create stuff that belongs to the game space; but when you take a closer look, it seems to be much than a release of commercial design software (such as CAD, etc.)
Recently, I’ve been looking at virtual environments that are becoming more and more popular, in my opinion. The first well known one is Second Life developed by Linden Lab. But recently, I have seen few new environments. One of them was developed by a company – Metaplace
What is interesting about this software is that it is very similar to blog-following software, as it works inside of a browser.
From the customer ‘s standpoint, this functionality is very useful in scenarios relating to collaborative design and social experience. What if PLM companies were to use these capabilities to develop design software that would run or integrate with these virtual worlds? Design would therefore become social. Another possibility would be to integrate collaboration capabilities that allow you see social groups and other contractual work in this virtual environment.
Posted by olegshilovitsky 
