CATIA V5 Migration Challenges for Large Enterprises
by Roy K. Rajan, Ph.D., Chief Architect, CAD Migration, Lockheed Martin Aeronautics Company

Large enterprises are facing a series of challenges in migrating to CATIA Version 5 (V5) from Version 4. Strategic, as well as cost elements, have to be taken into account and optimized with technical capabilities to achieve a successful migration and end state. The challenges are multi-dimensional and involve complex scenarios that are inter-linked.

Heritage Companies & Legacy Programs
The CATIA landscape at LM Aero is dotted with heritage companies, and legacy programs. The opportunity for V5 covers a broad spectrum from clean-sheet to drawing based programs that were initiated before computer aided drafting applications were invented. The technical processes span from fully life-cycle managed scenarios to concept explorations involving only a handful of specialists for short durations in isolated environments.

Fail to Plan and Plan to Fail
The current state of V5R5 to support specific scenarios and a forecast of the technology curve for future releases are minimum pre-requisites to initiate V5 migration. Product development roadmaps, release planning, industrialization processes, third party product strategy, lessons learned from other companies implementing V5, and external resources can assist in forecasting the complexity, span and resources for a control plan for the migration. Expectations of the end relative to the starting state should be managed for each program and operating unit.

Push and Pull
Introduction of new technologies adds elements of risk and cost for the Managers in tightly controlled programs. Technical champions and super specialists demand new tools to improve the throughput and quality of their work. Vendors and partners in the value chain push for new technology to minimize operating costs. A push-pull approach drives innovation while minimizing risk and cost in an extended enterprise.

State Transformation
CATIA V4 is used in several disciplines in large enterprises. Each of these disciplines represents a set of scenarios that are candidate for migration. A clear understanding of CATIA usage in the enterprise and specific scenarios in programs is a baseline for planning the transformation to V5. Showcase programs, opportunities, and investment decisions in the corporation typically drive V5 implementation. The base of scenarios can also be used as a gauge for measuring the progress of the migration. A discovery phase combined with pilots for risk reduction and validation are key elements for the thorough migration of all CATIA disciplines. Due to the extended nature of the enterprise, a big-bang approach is usually not practical. Phased implementation is usually the norm based on domain groups with less cross-connections and highest returns.

Migrating Up Stream
Traditionally new technologies are introduced downstream and migrated upstream to minimize disruptions. Consumer groups of CATIA data have to be planned upfront in the migration like planning, manufacturing and the extended supply chain for design/build programs. Concept exploration programs have more degrees of freedom for the producer groups and can server as early pilots to validate critical path design/analysis scenarios before committing revenue-generating programs to V5. V4/V5 hybrid scenarios for migration should avoid bi-directional data exchange scenarios to minimize data loss and re-work cost.

Attack of the Killer PC's
The introduction of V5 on PC's presents the opportunity to lower total cost of ownership (TCO) for the IT infrastructure. Large enterprises have significant UNIX investments in place to support V4. PC's were traditionally relegated for office applications while UNIX clients/servers were optimized for computer based design and analysis work. The merging of the two worlds presents opportunities as well as challenges for CATIA administrators and related IT organizations. Multiple solutions exist to synchronize V5 on UNIX, and PC, respect program boundaries for data, security control, remote distribution, user authentication across the two worlds, performance and availability. A discovery phase is essential to select the correct solution.

A Custom Fit
Most large enterprises have also customized V4 for process compliance, productivity improvements and functionality enhancements. V5 presents an opportunity and challenge to re-baseline the customizations that are must haves, should haves and nice to haves. Customizations have a hidden cost in terms of maintenance, and certification for new releases and service packs. A learning curve of 12 months is not unusual to transform a V4 programming specialist to V5.

V5 or Bust
It is not unusual to find a number of 3rd party applications in large programs in specific roles for select scenarios. They inter-operate with CATIA for application and data resources. Traditionally the vendors of these applications present a delay in migrating to the newest release. The core sets of applications need to be identified and the vendors must be prompted to achieve the proper level of competence and certification on V5. The options for satellite applications that are not part of the core set should be kept open.

Operational Support
Introduction of V5 impacts several organizations in the enterprise. The lead times for all the resources from the respective organizations have to be taken into account in the forecasting cycles to achieve full service readiness. It is common to miss delta infrastructure components, spikes for support needs during the introduction and a healthy reserve to manage unplanned events after the implementation.

United Front
Partnerships with Dassault, IBM, and other companies who have implemented V5 can accelerate mission success by sharing lessons learned, and serving as incubators and laboratories for innovative ideas. Commitment by programs, functional organizations, stakeholders, service providers, Dassault, and IBM are key milestones before implementation plans are executed.

The Big Picture
V5 migration must be treated as more than a version change. As everyone agrees, a process change is requisite to achieve a high return on investment for the migration costs. The technical dimension is only one aspect for a successful migration and should be balanced with strategic plans and investment constraints. Various degrees of success in V3 to V4 migration should be a reminder for most enterprises.