CATIA V5 and Enterprise Integration at Pratt & Whitney Canada

Pratt & Whitney Canada is a subsidiary of United Technologies, a U.S. multinational involved in various high-tech sectors. United Technologies provides high technology products and services to the aerospace and building systems industries worldwide. UTC companies are industry leaders and include: Pratt & Whitney, Sikorsky, Hamilton Sunstrand, Carrier, Otis, Chubb and Fuel Cells International.

Within the Pratt & Whitney Group, Pratt & Whitney Canada (PWC) has a world mandate to design, develop, manufacture, market and support turbofan, turboprop and turboshaft engines  in a wide range of power ratings to meet the diverse needs of the marketplace - regional, business, utility and military aircraft, and helicopters, as well as turbines for auxiliary power units and industrial applications.

In 2001 PWC undertook a multi-phase PLM effort that included the migration to CATIA V5. The reasons behind this move were numerous. On the business side, the expanding global environment was driving the need for instantaneous access to all product related data from anywhere at any time. With major engineering design offices in Montreal and Toronto and satellite offices in Russia, Poland and India, seamless access to systems and data is critical.

For our manufacturing activities we need to support plants and suppliers across Canada, as well as in the U.S., Europe and Asia. Our aftermarket activities span the globe with our repair centers located around the world – right where our customers need them.

All of these factors drove the need to re-evaluate our technical systems. Pratt & Whitney Canada was not new to the idea of integrated systems. In 1999 we successfully implemented SAP R/3 in a ‘big-bang' approach that resulted in the elimination of over 80 legacy systems.

We view the SAP implementation as covering our product execution processes and it handles all the transactional work related to ERP, CRM and SCM. We view PLM as the systems and processes necessary to integrate the product development activities and for this we are using the Dassault Systemes suite of tools. This concept can be viewed graphically as follows:

The linking of these two major process streams (ERP & PLM) is critical and it is this higher level of interaction that we plan to achieve. We refer to this as the "Digital Enterprise" and our goal is to become the first aerospace company in the industry developing engines using digital technology.

An important first step for us was to make the move from CATIA V4 to CATIA V5. The process of doing this started late 2000/early 2001. As with most of our large system implementations there was really a two-pronged approach to getting started – a bottoms-up grassroots effort and a top-down ‘selling' of a strategic vision. Unlike our product execution systems (integrated through SAP), our technical systems were still a mix of technical platforms (UNIX and PC) and a combination of commercial and in-house developed code. We were also already using Dassault Systemes products such as CATIA V4 for our single CAD/CAM system, VPM for vaulting our graphical data and ENOVIApm as a market feedback tool and to request engineering changes.

The grassroots level evaluation of CATIA V5 was done to determine if the software was ready, if there was enough new functionality to justify the move and to see where our processes could be enhanced through the use of this new software. A formal fit/gap analysis document was created and the decision was reached that it was time to take a serious look at how and when to move forward.

At the same time we had to ‘sell' the idea across the company. It took a solid business case and meetings between senior executives at PWC and Dassault Systemes to accomplish this. It was clear that our two companies had complimentary visions of the future. Even with all this, there was still what can only be viewed as a ‘leap of faith' by key project sponsors that gave us the final push and resulted in approval of the funds in early 2002.

We started with an exercise to generate a high-level ‘end-to-end' mapping of our business processes and identified where we had technology gaps. From this we defined project phases and timing;

Phase I (2002 – 2003)
CATIA V5, VPM, ENOVIA Portal
PLM study
FEM fit/gap analysis
UNIX to PC for Design & Drafting

Phase II (2004 – 2006)
Implement PLM solution
Implement FEM functionality

Phase III
Inter-company collaboration/integration
Data security

Our project spanned our Engineering and Manufacturing organizations and we put a project structure in place that included resources from PWC, IBM, Dassault Systemes, CSC (our IT outsourcer), and local companies with strong CATIA skills. There was strong executive sponsorship and program management that spanned all project and organization lines. We had an on-site Dassault Systemes assignee and we view this as a critical factor in our success as we needed and received strong support from the labs in France during the project.

We developed a methodology around the project implementation that has helped keep us focused and on track. We did not do a ‘big-bang' implementation but rather a progressive roll-out. Our engineering organization is structured by Module Centers, or major sections of the engine such as compressors, combustion, turbines, etc, and this is how we planned the deployment of CATIA V5. In manufacturing it was by part families.

For each roll-out the team defined a clear scope of what was to be done and assigned roles and responsibilities to each team member. For V5 for example, the team structure was well defined with resources looking at: admin, interoperability, solution development, training & roll-out, methods development and the V5 environment.

Aside from the core team, who stayed through the project, users from each of the areas were brought in full time for approximately 3 months to learn the new software and redefine the work methods for their area. The core V5 team mentored these ‘power users' to bring them up to speed quickly and share lessons learned across groups. This was done in a lab environment, disconnected from the production network so there would be no issues with experimenting with new functions. This network of power users remains in place today and they are called upon to test new releases and act as first line support to their groups.

For each area a consistent project methodology was followed to first map the 'as-is' process in the area, map the ‘to-be' or desired future state and then identify any gaps. While any required development was done, the power users returned to their jobs. They came back into the lab to validate the final solution. Once the power users ‘signed off' – and this was a literal sign-off and acceptance – the training started for the rest of their group. Users were given 5 days of basic CATIA V5 training in groups of 10 to 12 resources. In both engineering and manufacturing some users received additional specialized training.

While the users were on training the technicians got busy. UNIX workstations were removed and replaced with a PC fully loaded with CATIA V5 as well as standard office software tools. The mail system was changed from Netscape mail to Lotus Notes so that we finally had one mail system across the entire company. On the Monday morning when they returned from training the users had a completely new work environment. That's when the fun really started!

We underestimated the need for training on the basics of the new environment for the first roll-outs. We made an assumption that everyone knew how to use a PC and just cut them lose. This was not the case. A home PC is one thing but we changed the whole environment on these people. Basic things like the differences in storing personal files between our UNIX and PC networks were not clearly explained. We learned and adapted and did a better job on the next groups.

Just to put things in perspective here are a few facts. In January 2002 we had 1089 UNIX based CATIA V5 seats and 26 PCs in a development and test status. By December 2003 we will have 700 CATIA V5 PCs and 347 UNIX seats. Our first group was roll-out in May 2002 on CATIA V5 R8. We are currently running R10sp5 and are testing and planning the roll-out for R12sp2. There has been a lot of change in a very short time!

So why have we put ourselves through this? We did it because we saw the benefits. Moving from UNIX to a PC platform reduced our costs but also opened the door to more integration with our supply base as they will now have an easier time using the same tools we do. CATIA V5 brings new functionality, supports new work methods and gives us productivity gains we couldn't get with V4 through things like parametric modeling, use of power copy and rules based design. The V5 platform itself allows better integration with our other tools.

In manufacturing CATIA V5, coupled with some in-house development, allows faster NC programming time, faster machining time through improved definition of tool paths, easier integration of design changes and reuse of information across families of parts.

The Human Modeling software in V5 is helping us with virtual modeling of our engines for maintenance simulation purposes. This means that we can show a customer, right up front in design, that our engines can be maintained in the field. This is a big customer satisfaction gain for us. Clash detection and analysis in V5 is allowing us to find any potential issues right up front.  

We have linked our VPM installation to CATIA V5 through the ENOVIA Portal and our future plans include using the portal technology with our remote sites as well as our customers and suppliers. This will allow us to maintain one central VPM instance and avoid the need for hardware and replication of data around the world. These are important cost avoidance savings.

So what did we learn through this process? We learned there are a whole lot of common sense activities that need to get done – but you still have to do them! Here are just a few points......

• Pick your project team carefully. You need experienced people who are well respected and credible to your organization. You are going to be asking people to make fundamental changes to the way they work and the people asking need to be listened to. Support your team through the highs and lows of the project.
• This is a BIG change and trying to pretend it isn't won't change that fact. Plan carefully but be willing to adapt along the way.
• Train on the core products but also on the environment. Our move from UNIX to PC was bigger than we thought. And remember, a learning curve exists – don't try to pretend it doesn't.
• Communicate ten times more than you think is necessary. We thought we had covered this well with meetings, documents on the Intranet, newsletters, etc but we still got told "no one" knew what we were doing. Listen when you hear these things and adapt your style of communication to what the people want.
• Once you've deployed an area it's not over. Plan for a stabilization period and maintain a level of post-implementation communication & support. Keep a strong link with the power users. You'll need them for future testing and they can let you know if there are issues.
• Capture the benefits of the project as you go otherwise you have a moving baseline. Since we are implementing PLM in phases we need to prove we are getting the benefits we projected so we get the funding needed for the next step.
• Keep your executive sponsors well informed & involved. The sponsors are there to remove roadblocks and resolve issues that the project team can't.
• We had a big IT component to our project since we were changing platforms and also updating our infrastructure at the same time. We needed a strong partnership between the business units and IT to accomplish this.

The roll-outs are almost complete, now all we have to do is Phase II and Phase III......