PLM at Jaguar Land Rover – The Moment of Truth for Dassault’s 3DEXPERIENCE Platform

Sometimes crisis can bring about surprisingly positive effects. When existential questions and issues of survival arise, they can sharpen the mind and bring forward hidden talents. This can happen at an individual or company level. When viewed from this perspective, crisis can be a catalyst for success.

In 2008 Jaguar Land Rover (JLR) - then owned by Ford - was on its knees before being saved at the last minute by Tata Motors of India. This transaction followed right on the heels of the crash of the investment bank Lehman Brothers, at a time when virtually every car maker in the world had to fight for their existence. Ford was no exception, and so Tata bought JLR while Geely of China acquired Volvo Cars (VCC), another of Ford’s premium brands.

Today both companies are financially healthy, sales are growing and they’re at the forefront of the car and SUV industries. Crisis really did trigger success. However, it has been a tough journey.

From a PLM standpoint, this is an interesting story since both of the brands were using Ford’s C3PNG program. Thus, they started from the same PLM starting point, but have since gone in very different directions.

Volvo
VCC kept and further developed its Ford heritage (mainly Dassault Systèmes’ CATIA V5 and Siemens PLM Teamcenter in combination with its proprietary product database and configurator KDP).

Jaguar Land Rover
JLR took another route when it formally decided to migrate to Dassault Systèmes’ V6-architecture in 2010, and then on to the visionary 3DEXPERIENCE ”beyond PLM” concept. JLR was exploring the idea of a full Dassault-based venture, while at the same time evaluating ”the big three,” Siemens PLM (Teamcenter), PTC (Windchill) and Dassault.

Siemens was eliminated in JLR’s evaluation – 7 years later, it still uses Teamcenter

Siemens’ Teamcenter (TC) was eliminated as an option due to architectural reasons. TC’s unified architecture wasn’t ready at the time.

But here is the paradox; Teamcenter is still a major part of JLR’s product data management backbone, generally used in combination with CATIA V5. Even in the last few years, JLR has bought new TC licenses in anticipation of the new PLM project.

In 2010, JLR started working on a platform program called iPLM, based on Dassault’s 3DEXPERIENCE. The platform covers the entire vehicle definition across 14 domains, including Bill of Materials, parts and assembly, requirements and verification. It will also be able to deal with electrical design, styling and CAE capabilities like multiphysics and FEA as well as embedded software.

Initially, the iPLM program was slated to start production during the spring of 2014. PLM platforms are notoriously hard to build and so, seven years after the separation from Ford, the Dassault software-based platform isn’t yet in full production. The goal is for it to be in place by the end of this year.

JLR is the Reference Case for Dassault’s 3DEXPERIENCE

To date, the automotive OEM market has been hesitant to bet on 3DEXPERIENCE. Sales have been sluggish – a fact that makes the JLR implementation of the iPLM-program quite significant. By the end of this year (according to the latest plan), it will reach a state of full implementation at JLR, becoming Dassault’s posterchild for its presence in the automotive industry. Of all the Dassault product development tools, the automotive industry prefers to use CATIA V5 and, to some extent, CATIA V6. To date the main players have shown reluctance to bet on the 3DEXPERIENCE product data management backbone, ENOVIA V6.

In light of this, JLR’s integration of 3DEXPERIENCE is the moment of truth for Dassault, as it is for JLR’s iPLM. What were the reasons for the bold JLR iPLM venture? Why did it have to take seven years to get the solution in full production? Can it give JLR a competitive edge and fulfill its aggressive business plan?

The migration challenge: 600 silos of information

What doesn’t kill you makes you stronger. JLR’s CIO, Jeremy Vincent, knows a lot about this. He came on board in July 2008 and has been responsible for the company’s major IT projects ever since. Today, this includes the pillars of IT in JLR: The iPLM project (3DEXPERIENCE) and two SAP powered ERP projects (TURBO/business processes and definitions, and a common global business template).

From the day Vincent took the CIO helm, the IT goal became one of transforming this iconic car maker, with models like Jaguar Mark II, E-Type and Land Rover Defender from a complex, siloed organization to a modern, standards-based one. The PLM platform had to have the ability to work as an end-to-end solution rather than a data management system.

Judging by sales, this transformation worked well:

• 2012: JLR sold 375,000 cars (Land Rovers accounting for the bulk of sales)
• 2013: sold a record 425,000 vehicles
• 2014: showed another record year, with almost 463,000 vehicles, up nine percent from the previous year

This is a streak that JLR doesn’t want to break. Product development tools, including PLM are one of the crucial enablers, not only to get the technical side of product development together, but also to connect product development to sales and ultimately to its customers. It’s not a surprise that this ambitious program has turned out to be harder and more time consuming than anticipated. JLR’s Ford inheritance left it with a complex IT infrastructure.

”Our IT history has generated more than 600 silos of legacy data”, says JLR’s Paul Davies, director of Product Development Operations.’ The majority is not intercompatible. To flourish in business, all our data must be interoperably integrated and intuitively available to every JLR stakeholder”.

In a perfect world, that kind of ambition is feasible. But in an organization with more than 8,000 employees, like JLR’s, it has its limitations. Gartner’s PLM analyst Marc Halpern has seen this before.

”While PLM evangelists are enthusiastic about making significant shifts in processes/practices/organizations/applications/application architectures/data architectures to enable the next generation of product design and manufacturing, engineering and manufacturing organizations will take a long time to learn, accept, and adapt to the changes that must be made,” he said as a general observation on the challenges that the automotive industry is facing.

Such adaptations takes time and the scale of change is a problem. Change has not always been well received in parts of the JLR user community.

"We are still on the journey of changing perceptions of IT and there will always be supporters and detractors,” Jeremy Vincent said to CIO online magazine in late November last year. ”Some of my stakeholders say 'I just want a system.' But a system is technology, people and process. We are still mopping up some of the shadow IT, and shadow IT is still being created."

He added that the iPLM development was more than halfway through and explained that a major obstacle is tied in with legacy models, i.e. migration of product data from old, non-compatible files from old systems.

JLR’s ”tactical strategy” towards a fully functional iPLM platform by 2015-16

The issues of compatibility and migration are, however, old news at JLR. As early as 2011, when I conducted an interview with Paul Davies, he pointed this out. The company had been working with the problem for a couple of years, he asserted, and it had a solution in sight.

”Look, we had a very fragmented and disjointed PLM landscape,” said Davies. ”JLR is in an engineering and creation industry, so for us to transform our business we have to alter the creation process and PLM is the fundamental enabler. So what we have done is to deploy a number of what we call ”Zero pilots,” which is where we’ve taken a tactical area and said, ’we have some tool capability that isn’t working for us,’ but we also want to make some process changes and things like that. So for example, we use (ENOVIA’s) Program Central to help us replace our total program management tool that we use for managing that £1,5 billion worth of investment every year.”

The users like the interaction of it, he claimed, and said that it’s rapidly grown ”so much that we’re now getting demand to exit completely the previous tool, while we’re still running through some programs to try and migrate it into the latest tool.”

This was said in November 2011. Talking about the next steps, Davies explained that JLR was going to continue with the ”tactical strategy,” where it could build the solutions ”whilst we’re assembling the underlying backbone of PLM.”

Configuration driven BOMs

According to the original plan, based on the 2010 launch of the iPLM concept, JLR scheduled the implementation for the spring of 2014. This has been postponed until the year-end of 2015 or early 2016.

During 2015 the goal is to establish the fully functional platform through a gradual entry process containing four parts (”P1-P4”), which in turn will make it possible to develop and realize the first vehicle ever (”Vehicle 1”) on the completed iPLM platform.With these platforms in place, iPLM lead at JLR, John Knight-Gregson, claims that ”after four years in [the] making,” it will have the ability to execute on:

• Milestone driven configuration
• Configuration driven BOMs
• BOP/BOM/BOI/etc driven CAD
• Integrated Change Management

Here are the milestones of this journey from 2015 and onwards into full iPLM implementation:
MARCH 2015: Official launch date for Platform 1 Vehicle 1. MAY 2015: Reached Gateway 4C, which will allow JLR to start feeding the downstream systems directly from the 3DEXPERIENCE platform. In turn, this will mean that it is ready to start building cars from the Bill of Materials within Enovia (V6).
JULY 2015: Platform 2 Vehicle 1.
SEPTEMBER 2015: Issue Management - JLR is set to replace the Risk and Issues Management System across its operations, which means that this entire environment will be moved onto the platform.
JULY-SEPTEMBER (Q3) 2015: Platform 3 Vehicle 1.
DECEMBER 2015 Platform 4 Vehicle 1. JLR aims to effectively implement the new iPLM technology on every platform and vehicle by the end of this year.

”We don’t want to be in that place too long, so there’s a rather ambitious and maybe glib statement that we’re going to do the rest of the company over Christmas of this year. So, by early next year there will be about 17,000 people across JLR, and across the supply community that are on the iPLM platform,” said Knight-Gregson, whom I met in Munich during Dassault’s recent 3D Excite event.

A daring move to a rapidly evolving PLM platform

No doubt the decision to make an all-in bet on V6 and the 3DEXPERIENCE was a daring move. But since 2010, V6 and the 3DEXPERIENCE platform have evolved from CEO Bernard Charles’ vision to a reality with clearer contours and several seamless PLM functions.

The main applications on the platform are ENOVIA (PDM/PLM backbone), CATIA (CAD), SIMULIA (CAE/simulation), 3DVIA (lightweight visualization), NETVIBES (to gather information from social media regarding what people think about JLR) and DELMIA (digital production planning).

According to John Knight-Gregson, these apps (together with 3DEXCITE) are also key elements of JLR’s iPLM platform.

“With us separated from Ford we almost became a green field site”, Knight-Gregson said. “However, for the first time for us, we have genuinely started with a milestone driven configuration platform. So, everything we do in the company is driven from a plan, and a need to do something by a certain date.”

The objective of this milestone strategy is to control the configuration or the effectivity within the product. “That configuration then drives the Bill of Materials, or the Bill of Service, or the Bill of Process, within the company. And then the Bill of Process and the Bill of Materials drives the CAD, and effectively gives the geometry somewhere to be hung. And around that, for the first time, we have an integrated Change Management System which means that because we have a single source of truth, we can allow people to manage the changes and have complete visibility of every aspect of the change.”

As always in today’s complex product realization in the OEM automotive industry the BOM issue is a key to success. What JLR is aiming for in this context is not only difficult to carry through, but should it succeed the company would achieve one of the most advanced and progressive solutions in the world.

Dassault’s data driven environment in ENOVIA requires new ways of working

I discussed the general BOM challenges with ENOVIA’s CEO, Andy Kalambi. The background is that to benefit from the “Zero error targeted BOM,” which is one of the promises of ENOVIA V6, you have to work in a data driven, non file-based environment. This can cause problems since the overwhelming majority of Dassault V5 customers (JLR is one of them), OEM suppliers and others generally work within file-based environments (as is Dassault’s PLM V5).

The data-driven environment also requires a new way of working. For JLR, this has (according to my engineering sources in the company) caused some problems and also been met with some resistance among the staff.

Another issue is that CAD is not the only information being managed in an enterprise wide solution. A lot of the information from other ”BOM sources” that belong to the complete product lifecycle comes from file-based systems like Microsoft Word, programs for software design and others. “But in V6 we have solutions that can deal with this in an automated fashion,” said Kalambi

So, what are the advantages of the data-driven environment?

”When you are in a file based world – let’s say you’re using file based CAD – to extract that information and populate a BOM, you have to be able to manage a BOM by connecting to the CAD, automatically generating the part information and populating the BOM. However, the problem with that environment is that the part information is still sitting in a document, it is not in a database. So there’s always that time delay as the document is reconciled with the data structure”, Kalambi explained, and continued: “In a fully data driven environment, we’re not talking about that process. At any time you can make a real-time digital mockup, and your CAD and your BOM are completely lined up, which will never happen in a file based system.” The BOM should be generated, not created, he concluded.

A big deal for Jaguar Land Rover

These capabilities are a big deal for Jaguar, asserts John Knight-Gregson. He said that within the iPLM program a single feature code-set is now driving the Bill of Materials. “This gives us the ability to set up structured control models across the company, which are then consumed by all parts of the organization. And we have abilities to hang all of the digital information connected to the product into the BOM, such as color and weight; things that you wouldn’t traditionally think of being in the Bill of Materials, and cascading the requirements for the product into the specific part. This way the part designer knows what the requirements are for the engineering of that part.”

The visions that drive new technologies

Here’s my take: Without visions, there are no new technologies. Without new technologies, there are no amazing products or services that can change the world. Visions are the drivers of development and the ideas that Dassault’s Bernard Charles presented have affected the entire PLM industry and how products are made.

But that doesn’t mean that the implementation of those new ideas is problem-free. On the contrary, anyone who tries to move along the path of visions is asking for trouble. However, without these ideas, it’s difficult to move a company and its products to new levels of innovation.

The secret to success lies in the balance between realism, new ideas, education, capable partners/consultants and the pace of change. And most certainly the software has to deliver on its promise. When this happens, the breakthroughs almost always come slower than expected, but the reward down the line can potentially be huge.

Some of the main problems include obstacles such as how people deal with change, software modules that don’t work the way they were expected to, compatibility, steep learning curves, lack of best practices, etc. V6 and 3DEXPERIENCE requires a whole new way to work. The database-driven environment, as opposed to file-based, is just one example of that.

One of the reasons the PLM vendors have to continue to invent a new vision for the future is the requirement to support “Smart Car” concepts. Gartner’s Marc Halpern claims that the PLM software vendors are not yet prepared to support development of smart products.

”While they are all increasing the technology portfolios to support this (most notably Dassault Systemes, PTC, and Siemens), they do not yet have sufficient experience to wisely support their customers,” he said. ”Also, the vendors need to orchestrate the growing number of software assets in ways that best serve the customers. This will be a steep learning curve.”

Futhermore, the mix of skills needed to develop “smart products” is changing. The need for software engineers is growing faster than the need for mechanical and electrical engineers. In today’s cars embedded software accounts for more than 60 percent of the new vehicle development efforts. ”These shifts in the demographics of new product development organizations will have a significant effect on software vendor priorities to support them”, says Halpern.

This moving target of PLM capabilities can lead to self-imposed problems at customers. One of the drivers behind the lengthy project delays at JLR may be a desire at the company to continually evolve the vision of its ideal PLM implementation.

JLR has placed a huge bet on the 3DEXPERIENCE Platform

In the case of JLR, Dassault Systemes has been working hard, activating all of resources. JLR obviously made an all-in bet based in part on Bernard’s vision of the future. That gives Dassault a responsibility to live up to its promises. It can’t afford a failure. For its part, JLR has demonstrated a deep committment and even passion towards achieving this grand vision.

”iPLM is about PLM in the truest sense of the word; which is Product Lifecycle Management, and it’s about managing product data across the entire life of the product,” Knight-Gregson said. “Our view is that that digital asset should be consumable at every stage of the process. So the digital asset from design is equally important to virtual servicing, it’s equally important to CAE and it’s equally important to packaging. So that digital creation that happens early on, can pass right through the lifecycle of the car.”

One reason JLR will face challenges is because a transition of this scale hasn’t really been done before. In many ways, this project reminds me of the PDM projects in the late 1980s and early ‘90s: Nobody knew exactly what to do and it took at least 10 years to get the systems to function properly. It wouldn’t be surprising if history repeats itself.