Just in case you have not noticed, the world has gone mobile crazy. Mobile apps are now key business tools and Engineering is no exception. This raises an important problem: mobile by definition means that your data is going mobile; but how do we protect the Intellectual Property (IP) of this data? PLM holds data in the form of CAD Models, drawings and documentation, which is some of the most critical data that the company owns. Protecting IP is not new, for example, in 1967 Israel’s Mossad allegedly stole 3 tonnes of drawings of the Dassault Mirage fighter and effectively copied the aircraft, it was a huge operation; now someone can have 3 tonnes of drawings on their phone! Going mobile clearly poses a new set of problems.
Financially IP theft is not a small matter, the U.S. Commerce Department has estimated that intellectual property theft costs the economy more than $250 billion and 750,000 jobs annually and the International Chamber of Commerce estimates that the global fiscal loss is more than $600 billion per year. In another example, it is estimated that the worldwide turnover of fake automotive parts and components amounts to $12 billion a year, of which $3 billion is in the USA alone.
Protecting mobile data
Before we can protect our valuable data, we need to understand how it is used. Who needs access? What do they want to do with it? and where do they need the data?
The following are some examples of data going mobile:
- Managers making approvals whilst travelling
- Shop floor access to drawings and
- CAD models to check details
- Service technicians on-site repairing products
- Providing designs to OEMs and third parties to outsource manufacturing
- Design reviews with clients on-site
For our organisations to thrive the product data must be mobile, but how can we protect this data in the wild?
First we need our design data to be housed in a safe place and most companies have agreed that PLM is that place. All PLMs have access control mechanisms, usually a complex matrix of users, teams, folders and life cycles. Using rules we restrict user’s access to data. The first stage to protect data is to make sure it is carefully organised within the PLM. For example, in Windchill “containers” (aka “contexts”) were introduced some years ago to assist this this, and now form the backbone of the data and team organisation in the system and therefore the underlying access rules. However mobile adds a new dimension to access control, “If I want to see a drawing on a tablet, it is because I want to move it somewhere” and this needs us to have more that static data management rules.
In the Clouds
“Let’s keep it in the cloud, it will be safe!” this seems like a legitimate answer but it is not that simple. Unfortunately even if the data is stored in the cloud, the adage “If I can see it, I can copy it” applies. The data itself is fluid, it moves easily. For example, if I look at a drawing on a mobile devise a copy is downloaded, even if I try to prevent the user accessing the downloaded copy I can’t stop them doing a screen shot or even simply taking a photo with a high resolution camera.
The data is on the move, wherever it originated.
With so much at stake the industry has concentrated on locks, preventing unauthorized access to data that is mobile. A number of commercial solutions exist, to encrypt and password-protect files when they are viewed or downloaded, the viewer will attempt to limit the users’ ability to make changes, cut and paste, save etc. All the major CAD vendors have these types of solution. This software uses closed applications and proprietary file formats to limit access; perhaps the best known widely used closed application is Adobe’s Acrobat PDF viewer. Many companies use PDF to provide read-only access to drawings and documents but there are many other applications specifically for CAD data. It should be noted that however hard we try, we can never get past the “If I can see it, I can copy it” rule. They are many other techniques used to try to protect data.
Watermarking is a very active area of interest for many companies; a mark is added to a drawing which overlays additional information and in doing so makes it harder to copy the image. Using another approach some software providers have investigated Digital rights management (DRM), or more accurately IRM, but most seem to have rejected it as being too complex to administer.
Finally we need to consider the human factor, an Ibas survey (www.ibas.net) shows that only 28.2 percent of business professionals commonly think that intellectual property theft is completely unacceptable, and the most common thieves are the IT folks themselves, so maybe locks on data are not the only answer; at Wincom we are looking at other ideas.
The first is to make it hard work to copy the data. Many 2D drawing formats are vector based, meaning that the file is effectively as set of instructions on how to draw the drawing. This makes the result, small, fast and scalable, examples of this format are dwg and svg, however the problem is this format is very easy to copy, and even the watermarks are relatively easy to remove. Converting vector to raster when sending content to mobile devices makes it much harder to copy; albeit at the cost of larger file sizes.
Another technique we have adopted is called “personalised watermarking”. Wincom watermarks are applied at the moment a user views or downloads a drawing, and includes the name of the user and the time and date. This will encourage users to value and look after the data properly.
In addition to standard watermarks we also incorporate “hidden watermarks”. Once the data is in a raster format we can add information on the drawing that the human eye cannot see and is embedded into the data. This means that if the data is copied in any way, and then we get the hold of the copy in future we can identify who, how and when the data was copied. Having a leak is bad enough, not stopping it once you find it is worse.
For our clients we use a mobile PLM framework and a secure content server, which allows us to create custom task oriented apps, giving the user quick access to only the data they need to do their job and provides that data only in a secure format.
Accept the data cannot be 100% secure if it is mobile, but we can to make it hard to make unauthorised copies using raster formats, watermarking and closed applications. The next step is to get employees to have ownership and value the data they use. Keep data well organized and access control rules up-to-date. Finally give users access to only the data they need with task oriented mobile apps, which will make them more productive and reduces the risk of IP theft.
There is a relationship between PLM and ERP and therefore a need to integrate the two systems; at Wincom we create 4 or 5 of these integrations every year. I’d like to share our experiences with you.
ERP is all about efficiency, using the resources of the company to their maximum potential; PLM is all about managing creativity, ensuring that the innovation of a company is nurtured, but kept under control. Linking these systems together is a bit like a marriage, where each partner has their own personality and slightly different goals, and both need to work together to raise successful products. PLM systems have a tendency to want to over communicate, whereas ERP systems often think they don’t need to listen to the rather bothersome PLM. In fact, like any successful relationship, it is vital that there is good communication between them.
Where to start…
It seems that we need to connect the systems together, and if so, we need to answer some basic questions about this interaction:
1. How to send the data
2. What to send (and receive)
3. When to send the data
Before answering these questions, we must realize that the relationship between the two will take time to be established and will evolve. We need to begin with basic information, transferred as it is needed. Almost always, we need to send product data from PLM to ERP but later as the relationship matures we need to send data back from ERP to PLM. Parts and BoMs and information about availability, suppliers and costs can be vital information sent back to PLM to help the product designers.
How to send the data
The first problem is that the IT and ERP teams often expect that they will have large amounts of transactional data; this drives their technology decisions which then often tend towards the most mature data transfer technologies available (e.g., Tibco) that were primarily developed for banking applications. We have often seen heavy, expensive middleware solutions implemented at great expense that were designed for high volume secure banking transactions. But, is this what we really need? How often does the product get released or changed? Do I need to transfer this data every 10 milliseconds? Clearly the answer is no.
Recently the technology landscape has changed and now the ubiquitous web service is often put forward as a solution, but even this involves a degree of complexity. The place most companies actually start from is the basic idea of a single file exchange. Frankly why not? This basic approach helps to create a common vocabulary and put in place working practices that can evolve into “proper” interfaces. XML is a good way to format the data, but at Wincom we are often asked to use the humble comma separated file (CSV). (We even had one company that told us to create XML and then they quietly transformed it into CSV as that is the only thing their ERP developers actually understood)
What to send
So we don’t need a high speed middleware bus, but we do need to send accurate and sensible information. To begin with the ERP people need to understand a little about the PLM obsession with history (a good PLM is like an elephant, it never forgets) and the PLM team needs to understand the ERP focus on the here and now.
A good understanding of revision schemes, effectivity and change models all become important. It is interesting to note that the same words can mean different things to different people, so nothing should be assumed. Finally, it should also be noted that a working, formal change process is a normal pre-requisite to an interface, as it is only then that the “creative” product data is under a sufficient level of control to be of practical use to the ERP.
With a common vocabulary in place, a formal declaration of the nature and format of the data to be exchanged can be agreed.
When to send the data
Finally “when” is important; the trigger to send the data usually comes from the change process of the PLM, often as a change reaches a certain level of maturity in its lifecycle. The user interface may also allow authorized users to trigger a transfer of data with a custom menu option. Tools are sometimes needed to trigger bulk transfers of data. Data may be transferred as little as once a day but it needs to be sent automatically and securely, with complete traceability. But, as we said before, we have no need for high volumes of data to be transferred instantaneously.
Conclusions and future
Once the two partners have started to talk and begin to exchange information, it is not the end of the story. There are huge benefits to be gained by expanding the way the systems communicate with each other, and also to get the ERP to open up and send data back to the PLM (such as supplier data). If we can have open, clear communication we can start to get real benefits and raise fit, strong and healthy products.
How can the extended enterprise leverage PLM?
These are real questions from real customers we have been asked in the last year.
– Our service engineers on site in Africa need to see the latest drawings.
– The shop floor needs easy access to the latest process plans.
– The purchasing manager wants to see if they should order 10 or 10,000 parts
– The engineering manager needs to approve a design whilst getting on a plane
A fully featured PLM such as Windchill or Ennovia has an interface that scares most people, even inside the engineering community. However we need this power to answer our complex engineering questions, but PLM also holds information that is crucial to the success of the extended enterprise, so how do we make it more accessible?
People use technology differently now; apps on mobile platforms are one clear example. “I don’t need a map unless I am studying geography; I want to get from A to B”. In PLM terms: “I don’t want to learn about effectivity; I want to find a drawing”. Simply put, an app is a highly focused interface that does one thing well, like find a drawing.
Interestingly this shift is happening everywhere, as pointed out by Oleg Shilovitsky, PLM Think Tank in his article “What Social PLM Can Learn From Facebook Decline?” where task oriented apps, such as messaging, are being used in preference to Facebook in some situations. Now Facebook is not about to go bust, but apps like WhatsApp and Line are definitely on the rise.
PLM is full of very useful information, but it is really used primarily by users who live and breathe engineering. A simple app does not compel users to enter the PLM, it gives them a limited view of the PLM, focused on what they need it for. “There’s an app for that”. A classic example would be a mobile or tablet app and each one is designed to help with a specific task, however apps in business should not just be mobile, they may also live on the desktop but they are always simple to use and highly focused on an activity. If you think this is another fad then think again, you probably already use a task oriented app, configurable reports are simply non-interactive “apps”
Maybe if we can deploy a few strategic task focused apps in our organization, we can provide the extended enterprise a vital source of information, without teaching them all PLM. (they don’t want to learn it anyway)
Vendor Mobile apps
The big mistake PLM vendors make here is to try to make a mini-PLM. Mobile is different, and the applications should be task oriented, is it really important to have a 3D model on my iPad (except because it looks cool)? Custom apps can help here, and the PLM eco-system is starting to create a number of apps that plug into PLM’s like Windchill.
For example an app was developed for a Bell Equipment that makes trucks, big trucks, which are used across the world. Bell wanted to put in the hands of their service engineers a tablet app that given a truck id, they could have instant access to the manuals and drawings. Interestingly a future update will be an offline mode, as the customer pointed out to us, they have tablets but there is no high speed internet in the jungle.
Searching outside the box
Searching for information is another classic app and there can be configurable apps that show different searches to different users. The PLM administrator configures searches, each one is for a different task, some via mobile and some using the desktop. Shop floor can find process plans, purchasing can check part usages etc. etc. Nobody has to access the PLM to find information, simple apps, quick answers.
Watch this space…
The world is not going to stop using powerful PLMs or that apps will replace them, but the way we interact with information is changing and engineering data is no different. Also it is getting dramatically easier to create apps and coupled with a PLM like Windchill with its open architecture there is a real return on a quite small investment. Technology is moving fast and if we can “think differently” maybe we can apply it to finally get PLM to make its promised impact on the extended enterprise.