The programming language to hack next PLM system

July 31, 2015

tech-crunch-hackaton

Have you heard about hackathons? According to wikipedia, hackathon or “hack days” is an event where software engineers including designers, project managers and everyone else collaborate actively on a project. It is also a term that used for more organized innovative effort. Hackaton organizations goes back to roots of open source software. I can see a growing interest in hackatons as an organization of a specific effort to create a new software or make a significant progress in a project collaboratively.

TechCrunch article Which programming languages get used most a hackathons? gives you an interesting outline of tools and services people are using these days to hack future systems and products. Navigate to the following link to get a full research list. Among platforms, programming languages, databases and communication frameworks I was a bit surprised to see Autodesk services (Well done Autodesk!). None of other providers of platforms, APIs and services from engineering world wasn’t mentioned. You can also see a growing number of hardware platforms, which gives you an indication of large number of hardware development.

The section related to APIs caught my special attention.I’ve been talking about Web APIs yesterday in my blog – PLM and spreadsheetware. Take a look on the picture below

Screen Shot 2015-07-31 at 9.25.06 AM

Communication APIs, social APIs, payment APIs, Geo APIs. This is where web programming environment is heading these days. Web APIs and RESTFul services. Combined with agile development methods it gives you a good foundation how to develop open systems. Do you remember my old post – Is PLM customization a data management titanic? For the last five years, CAD and PLM industry moved forward with development of SaaS applications and services. It is a time to turn SaaS applications into consumable web APIs and services that can be re-used to hack a better PLM experience.

What is my conclusion? The time when companies developed system for 2-3 years to bring them to the market is probably over. The idea of hackatons and agile development is compelling. It can be used for new product development and hacking of service based products to solve a specific problem. Manufacturing is going into networking future and no-stack software era. To provide a set of consumable open services can become extremely important. Just my thoughts…

Best, Oleg


PLM and Manufacturing in Networked World

July 21, 2015

plm-networked-world

Our dependence on networks in our life is growing every day. Look around you and you see networks everywhere. Just cut network cable in your house – chances are you are loosing most of your communication channels. But, our network dependencies are not limited to communication. Networks have more transformative influence on how we live, work and do business. It comes in a way of building network-related dependencies. I’ve been talking about some of them in my COFES 2015 presentation – Product Lifecycle, Supply Chain and Data Networks.

LinkedIn founder Reid Hoffman published a great article about future of driving in network age. The article is a bit long, but I recommend you to have a read. With examples of car and driving eco-system, the article shows the magnitude of changes network transformation can bring in the industry – improving highway throughput, reduce collisions, optimize parking process. The follow passage is my favorite:

Already, the car as network node is what drives apps like Waze, which uses smartphone GPS capabilities to crowd-source real-time traffic levels, road conditions, and even gas prices. But Waze still depends on humans to apprehend the information it generates. Autonomous vehicles, in contrast, will be able to generate, analyze, and act on information without human bottlenecks. And when thousands and then even millions of cars are connected in this way, new capabilities are going to emerge. The rate of innovation will accelerate – just as it did when we made the shift from standalone PCs to networked PCs.

…technologies that allow cars to talk to each other, through Wi-Fi-like networks that use dedicated short-range communications frequencies, exist too. In these vehicle-to-vehicle (V2V) networks, cars share information with each other and other smart infrastructure elements – traffic signals, sensor-embedded roads, roadside cameras, eye-in-the-sky traffic drones, etc.

The same networking paradigm will apply to other industry segments as well. For the last few decades, manufacturing became global with companies leveraging market, design, engineering and manufacturing facilities located around the globe. The growing specialization in specific manufacturing verticals created industry of contract manufacturers and suppliers. Many of them are acting like independent vehicles on the road today – following rules, infrastructure limitations and trying to optimize their own local path towards the goal. While cost are still top manufacturing concern, similar to single car on the road, to optimization is highly unpredictable. It happens to manufacturing companies acting independently as well as divisions of large manufacturing giants separated by siloed organization and software.

What is my conclusion? Manufacturing companies are going to have a lesson of networked world. It will be impossible to optimize the performance of single manufacturing entity without relevant network information. It will not happen overnight. Companies will try to gain more information about networked world around them similarly to how GPS developers gathered information from other cars. Intelligent PLM software with network mind can provide a competitive power to future manufacturing. Just my thoughts…

Best, Oleg

Image courtesy of KROMKRATHOG at FreeDigitalPhotos.net


How manufacturing companies can open source PLM innovation?

June 5, 2015

data-management-innovation

Whatever you can hear about complexity of PLM, there are lot of manufacturing companies that created outstanding PLM implementations. These implementations are helping them to optimize product design, speed up engineering to manufacturing process and do many other innovative things making their engineering and manufacturing uniquely competitive. From my experience, these implementations never came as a result of single PLM platform deployed in a company. It is usually a bunch of projects that are using several PDM / PLM products combined together with code and some homegrown stuff. In some situations these combinations are uniquely representing companies’ needs and ideas. I found it a gold mine of PLM implementations. But, very often, these implementations end up with a single company using it for their internal purposes only.

I think, this is a time for manufacturing industry to learn what other innovative companies are doing in such situations. Airbnb, the company helping people to rent apartments around the globe just made an announcement about open sourcing some portion of Airbnb data management infrastructure. Navigate to the following readwrite article to lean more – Airbnb Opens Data, Machine Learning Code. I found the following passage specially interesting

For Airbnb, which makes money by operating a marketplace for unconventional lodging in people’s homes and apartments, not by selling software, releasing its in-house software projects as open source serves to bolster its image as a technical innovator. There was a lot of talk about the company’s “engineering brand” at OpenAir. Allowing others to examine and use its code should, in theory, help the company recruit engineers and retain its current technical employees by publicizing their coding feats.

If you interested to learn more, navigate to airbnb.io – the destination of all Airbnb open source projects. If you are technical, you can find some very interesting things to experiment over the weekend. My special attention caught Airflow. I guess lot of my PLM friends can be interested by DAG-driven workflow scheduler.

Use Airflow to author workflows as directed acyclic graphs (DAGs) of tasks. The Airflow scheduler executes your tasks on an array of workers while following the specified dependencies. Rich command lines utilities makes performing complex surgeries on DAGs a snap. The rich user interface makes it easy to visualize pipelines running in production, monitor progress and troubleshoot issues when needed.

So, what about Open Source PLM you may ask? In the past Open Source PLM was a hot topic. Aras Corp. was the most active company after open source PLM innovation. Aras called it Enterprise Open Source. I don’t think the trend took off, actually. A subset of Aras solutions is available as an open source, but I’m not sure if these solutions can be used with another PLM backbone (which can be an interesting idea, btw).

The Airbnb example made me think about growing trend – innovative web companies are solving their data management problems without specific “enterprise software” vendors. My hunch we need to look more after what companies like Facebook, Twitter, LinkedIn, Airbnb, Uber, Tesla, Local MotorsI etc. are doing and how they are innovating with data.

What if some manufacturing companies will decide to open source their PLM implementations? Would it be something other manufacturing companies can leverage? Recently Tesla open source their patents to be used by other car companies. Maybe PLM implementation can follow the same path too?

What is my conclusion? PLM industry should find a way to innovate beyond the trajectory of large enterprise PLM vendors. The complexity of existing PLM platforms makes it dominant in the market and hard to compete with for smaller players. At the same time, manufacturing companies are suffering with slow ROI and complexity of solutions. Manufacturing companies can open source some of their implementations and create a dent in current big “top three” PLM dominance. Just my thoughts…

Best, Oleg

Image courtesy of KROMKRATHOG at FreeDigitalPhotos.net

 


3 reasons why big data is a big challenge for PLM

June 2, 2015

big-data-plm-challenge

Data was always a core part of what manufacturing does. Manufacturing companies have lots of data. According to Joanna Schloss of Dell Software, manufacturers are literally sitting on big data dynamite of potential revenues and opportunities driven by data initiatives. Joanna Schloss is subject matter expert on business intelligence and analytics; data warehousing, & big data analytics. Her recent article – On the cusp of a Big Data boom caught my attention this morning. According to her, there are several reasons that manufacturers can become a primary beneficiary of big data boom. The following passage can give you an idea why manufacturing can leverage big data:

Relative to other vertical markets, manufacturers enjoy three primary advantages that leave them uniquely positioned to benefit from big data. First and foremost, every industry and individual is touched by manufacturing in some way. You’re either doing business directly with a manufacturer, or purchasing something that at some point or another emanated from one.

In addition, because manufacturers were among the first to make widespread data collection a standard practice, they can quickly and easily scale their data collection efforts. Put more simply, a manufacturing company can track virtually everything much faster than most other companies can.

Lastly, manufacturers typically don’t face the data collection barriers that many other companies encounter. Whether they know it or not, many consumers readily provide valuable data to manufacturers on a daily basis.

The opportunity driven by big data can include improving product quality, help to discover new design for existing products and find new product opportunities. I agree with author – big data sounds like a gold mine for manufacturing companies.

It made me think how to bring these opportunity into reality. Do you think PLM vendors and platforms are in the position to make a play around big data opportunities? Manufacturing companies are sitting on piles of data. Existing business intelligence software was able to get this data, but wasn’t able to crank it until new big data technologies became available. I touched big data opportunity several times on my blog earlier – Will PLM vendors dig into big data? How PLM can ride big data trend in 2015; PLM… wait, Big data 2.0 is coming.

Big data solutions are quite unique in the way companies are implementing them. In my past publications, I was looking for examples of Big Data usage in product design, engineering and manufacturing. One of them was company True & Co that is using customer data to improve product design – PLM and big data driven design. Another example, I captured last month, is related to Siemens PLM big data projects based on Omneo platform. Read more about it here – Siemens PLM: cloud services and big data.

I’ve been thinking about the potential of CAD and PLM companies to leverage big data trend. My conclusion is that most of big data use cases are representing a big challenge for existing CAD/PLM vendors. Here is the summary of my thoughts. I can identify 3 main reasons for that.

1- Existing CAD / PLM systems are built on top of 15-20 years old RDBMS technologies. These platforms are providing limited capabilities to capture the amount and diversity of new data insight. Modern web and big data platforms are leveraging polyglot persistence principle that allows to use different database models to solve complex problems.

2- PLM platforms are built around the concept of closed world assumption (opposite to open world assumption) where all data models are predefined by a platform. Under open world assumption the data and statements about knowledge that are not included in or inferred from the knowledge explicitly recorded in the system may be considered unknown, rather than wrong or false. Existing PLM platforms have a big challenge to handle "unknown data" and be flexible enough to discover new data patterns.

3- The openness of PLM platforms are improving these days. A good example of that is Codex of PLM Openness focusing on how to establish data transparency between vendors, customers and services providers. Unfortunately, for most of PLM vendors, openness is reflected as an ability to export data from PLM system via predefined APIs. At the same time, it is hard to make design and PLM systems to be driven by the data coming from an outside world.

What is my conclusion? I think, big data is a big challenge for PLM vendors. Most of big data solutions are using platforms that are different and disconnected from existing PLM platforms built on older RDBMS technologies. Existing PLM platforms are suffering from limited ability to manage meaningful connections with big data platforms and are not capable to provide a platform to leverage big data insight and analysis. PLM vendors should discover how to apply modern data management principles to improve their ability to leverage piles of data and transform their solution from traditional data recording into data driven discovery and decision. Just my thoughts…

Best, Oleg

Image courtesy of jesadaphorn at FreeDigitalPhotos.net


Why PLM is failing to manage multi-disciplinary BOM?

May 21, 2015

high-tech-electronic

Products are getting complex these days. Look on every small electronic gadget in your hands. It is actually combined from multiple pieces – mechanical parts, plastics, electronic and software. Traditionally you are using separate tools to design these parts – MCAD, PCB design, software tools. Then it gets tricky a bit – you need to put together right information about the product, manage changes, coordinate with suppliers, etc. PLM tools are here to help. But, for some reasons, it is a difficult problem to handle.

Engineering.com article In High-Tech Electronics, Managing Three Lifecycles As One is a New Key to Product Development by Laila Hirr speaks exactly about that problem. Here is my favorite passage from the article explaining the problem:

HTE’s need for PLM is straightforward—a firmer grasp of the information generated before and during product development and subsequently “in the field.” Many information needs go unmet when products go into assembly operations of original equipment manufacturers (OEMs) and built into other manufacturers’ components in complex supply chains. Users and system integrators may also be slow to share information.

For many reasons, PLM has repeatedly fallen short in this industrial sector. At CIMdata, the reason we see most often is a lack of integration with the full information set that defines the product. Achieving this integration is a multidisciplinary challenge and in PLM’s twenty-plus year history with the high tech industry, the challenge has yet to be resolved. This largely accounts for the scarcity of compelling PLM successes in HTE and the ongoing skepticism about PLM.

Article speaks about absence of integration between tools and dependencies on homegrown spreadsheets to manage bill of materials and change. Which made me think about core problem in PLM tools – management of multi-disciplinary BOM. I addressed this problem in the keynote presentation at ProSTEP iViP Symposium few weeks ago – PLM and ERP: separated by a common Bill of Materials (BOM). PLM systems today are addressing BOM management. Most of them are taking an approach to manage multiple bill of materials view. However, these tools are not efficient enough to manage a BOM which contains mechanical, electronic and software pieces together. The complexity of BOM is driven by multiple disciplines, change management and product lifecycle as I presented on the following slide

bom-complexity-1

What is my conclusion? Technical difficulties and disagreement between people often can lead to problems in establishment of cohesive BOM management solutions. PLM fails to provide a way to manage multi-disciplinary BOM and changes. High-tech and electronic industry is specific because of high diversity of design tools – mechanical, electronic, software. PLM tools are not integrated well with design tool, which leads to poor BOM management. There are several reasons why it happens – limits of BOM management tools, complexity of integrations between design tools provided by multiple suppliers, UI complexity. Just my thoughts..

Best, Oleg

Image courtesy of Toa55 at FreeDigitalPhotos.net

 


What PLM industry can learn from… Hollywood?

May 13, 2015

PLM-hollywood

I’ve been reading a blog Getting PLM right: no one answer by Monica Schnitger. It is a slick writeup started from one of the most popular questions in PLM community – Why it makes my head hurt when I think I need/want/cloud use/maybe should think about PLM implementations? Monica speaks about why PLM creates so many problems that just thinking about that makes you feel bad. The core issue is simple – engineering and manufacturing is very individual. Each PLM project is different and it is hard to create or even think about “standard PLM implementation”. Although, I guess, some vendors will disagree with me and will try to sell you out-of-the-box PLM implementations. Here is the passage explaining the uniqueness of PLM implementations:

The concept of product lifecycle management, that you track many aspects of your product from inception through design to making to servicing to retiring/replacing — that’s common. After that, every case is unique. Also unique to each company contemplating PLM is its motivation. Do you have a specific quality problem? Need to better communicate with suppliers? Or are you trying to create a major corporate shift (“we want to be #1 or #2 in our industry”) and want to use PLM as a lever to re-examine old processes and shake things up? Each of these is a great reason for PLM, but everything downstream from that decision will be different.

A very nice video at the end of the article give you a perspective on modern trends in PLM implementations time and cost. Watch brilliant spoof by Jim Brown about new PLM tech called SprayON. Just spray PLM on your problem and it is over! Nice… More seriously, you can learn how modern cloud PLM products and technologies can create a new reality in PLM implementations in terms of effort, time and resources. To conclusion is simple – since cloud PLM can be instantaneously at your disposal, use it for a specific project first to get it done, show results, prove a value and move the next projects. This is great way to make PLM easier. However, PLM implementations are still hard and there are things that cloud PLM cannot do for you.

My thoughts about potentially magical new way to implementation PLM, took me to…. Hollywood. Have you heard about business approach called Hollywood Model in business? Navigate to the following NY Times magazine article – What Hollywood Can Teach Us About the Future of Work. The basic idea of Hollywood model is pretty simple. A project can be done by a team of professionals available on demand and assembled together for a specific period of time with the goal to make it happen and deliver the result. Here the passage, which explains that.

This approach to business is sometimes called the “Hollywood model.” A project is identified; a team is assembled; it works together for precisely as long as is needed to complete the task; then the team disbands. This short-­term, project-­based business structure is an alternative to the corporate model, in which capital is spent up front to build a business, which then hires workers for long-­term, open-­ended jobs that can last for years, even a lifetime. It’s also distinct from the Uber-­style “gig economy,” which is designed to take care of extremely short-­term tasks, manageable by one person, typically in less than a day.

With the Hollywood model, ad hoc teams carry out projects that are large and complex, requiring many different people with complementary skills. The Hollywood model is now used to build bridges, design apps or start restaurants. Many cosmetics companies assemble a temporary team of aestheticians and technical experts to develop new products, then hand off the actual production to a factory, which does have long-­term employees. (The big studios, actually, work the same way: While the production of the movie is done by temps, marketing and distribution are typically handled by professionals with long-­term jobs.)

Of course, there are lot of differences between movie production and manufacturing. But if you think more, you might think about it a bit differently. Especially if you look over the growing number of open source hardware projects, hardware startups, and other manufacturing initiatives. According to Ben Einstein of BoltVC in Boston, you can assemble hardware team of 8 people to create a new product. Read more here . The team will be working with many contractors in design, engineering, manufacturing and related fields to bring product to life. And remember, despite the fact hardware is still hard, there are many examples of small teams created hugely successful products.

In my earlier article Why PLM should revise NPI process?, I’ve been talking how future is shifting towards agile manufacturing processes and why PLM systems should revise the idea of structured workflow process as a fundamental approach to manage engineering and manufacturing process. I think, this is where future will take us. By the way, I didn’t find a hint on workflow processes in NY Times article about Hollywood production.

What is my conclusion? Cloud PLM is a step in the right direction. To remove a burden of hardware, installations and coding for customization is absolutely important. At the same time, it is tough to think about PLM as a life changing event for manufacturing company. Nobody likes the change and it is painful to get it (even with help of PLM consultants). But industry is changing. Manufacturing is shifting towards different organizational models. Mass production, large investments, big factories and long product lifecycle are going to be a thing in the past. Flat organizations, agile teams, pulling experience, ideas, skills, money and customers on demand – this is a way to think about future of manufacturing. The processes and software to manage it will change too. A note for software vendors, engineering IT managers and PLM practitioners… Just my thoughts…

Best, Oleg

picture credit NYTimes article. Illustration by Andrew Rae.


BOM and roadblocks for Product-as-a-Service in manufacturing

May 12, 2015

manufacturing-as-a-service

Software-as-a-Service (SaaS) is a software licensing and delivery model in which you are buying software subscription and software is typically centrally hosted. Here is a potentially big news – SaaS model is proliferating into manufacturing. Welcome Product-as-a-Service (don’t be confused with PaaS- Platform as a service). According to IDC report, manufacturing companies have seen some potential in after sales services and support. There is a good chance soon we will be buying services and not manufactured product. Here is a passage from IDC article to explain more.

By 2018, 40% of Top 100 discrete manufacturers and 20% of Top 100 process manufacturers will provide Product-as-a-Service platforms. Leading manufacturers have seen the potential that after-sales service revenues hold, with some generating up to 50% of their profits from after-sales sources. As manufacturers apply service innovation to their efforts, the product becomes a platform to deliver business outcomes and tangible value. IDC Manufacturing Insights defines product-as-a-service as the transformation of service from a standalone function within a manufacturing organization into an integrated product and service offering that delivers value in use. Manufacturers transition from selling physical products to selling the business outcomes the products will deliver.

Among the IT impacts we see as a result of product-as-a-service are the need for IT to support a global service delivery network with systems that enable the process flow for this new business model. Major systems, from customer management to service parts planning to finance and accounting will need to be altered and brought in alignment with how the product-service is brought to market. The change necessary is not to be underestimated, which is why we see approximately 40% of manufacturing reaching out to external IT service providers to assist with the implementation of product-service systems.

idc-manufacturing-innovation

To transform organization to sell services from selling products is a big deal. As it was mentioned above, it will bring significant challenges to transform existing IT systems. However, it made me think about challenges it can also bring to product lifecycle management environment and its core – BOM management.

The article Bill of Materials (BOM): Necessary or Just Nice to Have? from SparePartsKnowHow blog speaks about the role of Bill of Materials in services and spare part management. The article brings some interesting controversy around the need to have an up-to-date bill of materials in service. The article is very practical and I specially liked some examples. Here is my favorite passage:

If you choose to go down the path of software optimization (a theoretical approach) you probably do need to ensure that your BOMS are up to date. However, if you apply a pragmatic and process based approach such as the Inventory Cash Release process then the BOMS are less important. This type of approach forces you to look at the issues that drive your spare parts holdings.

These issues are not related to how many machines you have that use the part, they are almost universally related to the processes in place for spare parts management. These include: the basis of decision making (emotional, logical or data based), supplier relations, commercial arrangements, supply chain, procurement, planning, team behavior, and accountability. These are the most obvious examples.

From direct experience I can say that without a doubt these issues have far greater effect on your spare parts holdings than knowing whether you have X machines that use part Y. Of course the number of machines requiring a part will be an influence on the required holding levels, as it impacts demand, however, for the vast majority of companies their spare parts levels are far more influenced by the issues listed above and addressing these is the best approach for reviews and optimization.

That discussion reminded me very old disputes between Order Point and MRP strategies. It is clearly better to manage inventory by knowing what organization is manufacturing rather than maintaining a specific level of inventories. Moving into modern IoT era, I can see even more potential to correspond to a specific product requirements and needs to manage services and maintenance operation.

To have exact BOM of products in service can be tricky. This type of information is not well maintained by manufacturing organizations (especially, it is related to manufacturers that not using Serial Number BOM). It can be a challenge for this organization to move into more intelligent BOM management practices to bring up-to-date BOM in service management.

What is my conclusion? Manufacturing organizations will have to transform to support "Product-as-a-Service" model. It might create some significant IT challenges. One of potential challenges is the need to manage bill of materials for physical products in service and operation. The importance of BOM management will depend on specific "service management practices". We are going to see the evolution of these practices and related PLM technologies in coming years. Just my thoughts…

Best, Oleg

Image courtesy of pakorn at FreeDigitalPhotos.net


Follow

Get every new post delivered to your Inbox.

Join 284 other followers