Manufacturers utilizing control systemswith obsolete components are faced with the daunting task of needing to replace their system, even though it may meet their production needs and is very well understood by the operations team. What can be done to help improve the success of a control system migration?
Add More Value
The ROI of a migration project is sometimes difficult to justify. It is often viewed as an "insurance policy" against losing the ability to replace a "likely to fail" subcomponent. In many cases this type of justification gets postponed for many years but most likely there are features of a new system that can improve a migration's value.
Most systems that are obsolete have very slow storage components and little or no Ethernet network connectivity. It is very likely that the time to load recipes, update screens and process operator inputs is dramatically impacting an obsolete system’s yield or wasting valuable labor. We have some clients that report 20% improvement in yield just due to the responsiveness of a new control system's operator interface. If a control system can query its schedule or get product recipes from another corporate system instead of requiring manual entry, it is not only going to improve performance but also improve accuracy.
Changing the goal from "getting rid of the old" to "improving the system" changes the way a project is managed and treated.
Code Development Phases
When the obsolete system was installed it is likely that the hardware components of the system were 2 or 3 times the engineering cost. Control systems are now a fraction of the old costs and engineering tools have made configuration less tedious and more accurate.
Optimation finds that it is unwise to try to convert code. Instead it is usually best to plan on beginning the project as if it is a new system and use the old system as your "functional specification." In many cases, older code can be divided into a set of "common" strategies and rewritten following modular approaches better suited to newer systems. It is important that nothing is missed; going through each control loop, supervisory sequence, and operator screen is necessary to review it for functionality and/or determine if it needs to exist in the new system.
A clear description of new functionality that did not exist in the original system should be detailed. If possible, consider putting these new capabilities into place as a follow-on phase, even if that phase occurs closes after the initial migration. This type of approach allows the chance to verify that the new system meets the old functionally prior to changing the baseline.
Training and Simulation
Finally, it is likely that the operators will have a "learning curve" in moving to a new system. Training and simulation are the best way to reduce this impact on the project. Different approaches we have used include setting up the new system side by side with the old system and running dummy recipes in parallel with the existing system. This will verify that all required buttons, screen information, and reports match expectations. Some clients will have operators "pass" a test on a simulation before being approved to work on the new system. The danger of just setting operators loose without simulation and training is that they will become frustrated or make mistakes due to being disoriented.
Changing and modernizing is inevitable. Following a good plan that establishes goals, puts clear milestones in place and ensures a smooth transition will help a team embrace the change.