I had a couple of interesting discussions this week that made me reflect on something I’ve seen happen more times is ideal – the gap between a well-executed automation project and one that drags.
The insights aren’t rocket science (and probably universal principles around change to be frank), but automation stresses organizations in ways that other projects don’t.
That’s because when you automate a manufacturing site, you’re not touching one thing – you’re touching most of the ways the factory operates. Order processing. Queuing and batching. Inventory and picking. Manufacturing and shipping. And on top of that, you’re introducing new capability around agile production, transparency of effort, and order enhancement. Collectively that’s a lot of moving parts that are shifting all at once.
So, if you don’t have a clear view of how you will sequence and manage change – and therefore create the conditions for the project to succeed – the odds are that you either stuff it up or make it harder than it needs to be.
We’ve seen whole-of-factory transformations achieved in 16 weeks, idea to outcome. We’ve also seen the same scope take 52 weeks. Same technology. Very different outcomes. The difference almost always comes back to the same things.
6 principles
My 6 principles for running a good automation project:
- Clear air. Don’t start an automation implementation when you’re running three other initiatives at the same time. Create focus. Automation deserves its own window.
- Establish clear ownership (and know your real business rules). Don’t let executives lead the project. 95% of the time, they don’t know what’s actually going on – it’s hand-waving and principles. Production Managers and Supervisors know how the business really runs, including all the rules the front office has never heard of. Exec’s define objectives, teams execute.
- Be realistic about your order data. It’s probably dirtier than you think. Have your automation supplier rigorously test your order data before you go anywhere near go-live. The odds are it has more exceptions than anyone realises, and your team is already dancing around the daily dramas it creates – the front office just don’t realise how elastic production is.
- Pick a change pathway. Don’t try to change everything at once. Pick a logical grouping – brand, product type, plant – something that’s genuinely distinct. Implement the automation there first, prove it out, and grow from the success.
- Involve your people early. Don’t introduce the team to the change at the end of the project. Get them on the journey. Shared ownership creates belief in the outcome, and belief is what carries a project through the hard bits.
- Train and test properly. Don’t parachute the system in and hope for the best. I had a GM once who cancelled the training and testing phase and said, “give me the keys and let me drive the car.” Things got complicated real fast. Real change needs people who are familiar with the technology before it’s live – not after.
It’s frustrating (and expensive for both us and our client) to see projects get disrupted and delayed – especially when the root causes are avoidable. The technology is rarely the problem; it’s almost always missing context, poor direction, bad data or undocumented edge-cases.
If you’d like to have a chat about how to set your automation journey up for a smooth transition to better performance and profitability, reach out.
