Making Legacy Machines Smart

It’s a myth that the adoption of Industry 4.0 means getting rid of legacy machines.

To begin with, let’s define what "legacy machines" mean in this context. These are machines or industrial assets that do not intrinsically support real-time communication with a third-party system. In other words, they are not "smart machines." All industries that started a decade ago or more have a sizeable population of legacy machines. The older the industry is, the greater the number of legacy machines.

It’s impractical and prohibitively expensive to replace all legacy machines with smart ones. Hence, it becomes very important to integrate these machines into smart manufacturing systems. Otherwise, you will end up having blind spots in your plant. Business leaders should insist on integrating legacy machines into smart manufacturing systems.

The good news is that legacy machines can be converted to smart machines if you follow these steps:

1. Always start with business objectives. This includes identifying KPIs and what outputs are required from legacy machines to measure KPIs.
2. Identify a solution provider who has competencies to connect with legacy machines irrespective of their make and model.
3. Ensure the selected solution provider does a detailed survey of legacy machines. There can be multiple options for connectivity, as explained below.

The KPIs can be broadly classified in two buckets: production KPIs (OEE, loss analysis, and so on) and process KPIs (condition monitoring, diagnostics, predictive maintenance, and so on). For realizing production KPIs, at minimum, you need only cycle count and cycle time, plus some manual entries. For process KPIs, however, all the data measured during the production process is required, typically including parameters like various temperatures, pressures, flow, positions, and movement.

So what are the connectivity options?

1. Connect to machine PLC – this is the best option, and you can get both production and process data. Depending on the make and model of your PLC, direct communication may be possible, or you may have to add a communication module or a converter.
2. Install sensors – install sensors for getting production and process data.
3. Electrical signals – every machine, however old it may be, provides some electrical signals. These signals indicate events occurring during production, e.g., cycle start, cycle end, mold open, spindle running, and so on. These signals can be tapped as digital pulse outputs and used for getting production data from the machine.

An experienced solution provider will be able to survey your machines and suggest suitable connectivity options. There is a cost-benefit trade-off for each one. Ultimately, business objectives and returns should decide which option to choose.

So go ahead and make legacy machines smart!

This article is a product of the International Society of Automation (ISA) Smart Manufacturing & IIoT Division. If you are an ISA member and are interested in joining this division, please email info@isa.org.