Ten Top Tips to Deliver a Successful PI AF Project (Part 1)

By Steve Taylor, Principal Systems Engineer at ITI Group

PI AF Projects are often challenging because the equipment and plants they are built to monitor are themselves complex. The knowledge and data to model them exists somewhere, but it doesn’t yet exist in one place, and must be gathered, assembled and checked.

That’s a key benefit – yes, we’ll build metrics and KPIs to monitor and optimise plant equipment, but along the way we capture knowledge about equipment in one place, not just scattered across folders, datasheets and spreadsheets. When the next engineer joins, there is now a good functional model in place preserving knowledge about how the equipment operates and what to watch out for. There are dashboards in place showing them how their plant is doing, with cross-links to other applications they may need to jump to next. Getting to this point is the challenge.

Delivering a successful project is impossible without considering your strategy. Earlier in life, I was a huge fan of Strategy games – Command & Conquer, Age of Empires, Civilisation and various character and team-building games. As it turns out, whether you’re building a city, an army, or an industrial digital project, the game isn’t all that different.

Industrial Digital projects are tough, and by far the most common causes of failure (including runaway costs) are:

• Lack of business engagement
• Poor Subject Matter Expert availability to identify/clarify issues
• End user adoption failure.

All of these can be strongly mitigated by involving one or more specialists from your target user base, who are passionate about getting it right.

The most common mistake the digital project manager makes is believing that they are the hero of the game. If you’re overseeing the project, you are not the hero. One more time – YOU ARE NOT THE HERO. The hero in the project is the customer who will ultimately benefit from the project – your end users are the ones with dragons to slay, and need you to give them tools to finish the fight. They are Players 1, 2 and 3 – they will continue playing and re-playing the game long after you walk away.

If you don’t have one or more ‘heroes’ from your business to champion the project from inception to delivery, it will fail to deliver the best possible value in the shortest timeframe.

The casual gamer can get about halfway through a strategy game without very much strategy at all. You’ve built your empire and think it’s all going well, only to find the whole endeavour suddenly threatened by poor infrastructure: your roads are disconnected; your key buildings are too close together, or too far from the resources they require; you spent too much time or money on nonessential things.

In the real world we call this Technical Debt – projects reach their first, fifth, or final deployment, only to see servers struggle and die, or discover the design or deployment strategy does not scale. I have audited (and rescued!) many such projects: Technical Debt kills projects!

What you need in this moment is an expert – somebody who has played the game before.

Having at least one player on your team who knows the tech well is key – they help you set your architectural blueprint and shape your project correctly. They’ll help you get the hierarchy right, to build templates and displays that can grow and scale to the full needs of the project. The earlier this player is involved, the less technical debt you’ll need to pay for.

Don’t accept technical debt: get the basics right, and ensure you’ve got a clear path to scale up.

You’re preparing to slay a dragon – but which dragon? Is it a Firedrake? Or a Wyvern? Or a Wyrm? They’re all dragons, but can’t be defeated using the same weapons. Products have strengths and weaknesses, too – you can’t slay all your business dragons with a single weapon. The first question to ask is “what is PI AF really good at?”

1. Scale:

   If every single asset required unique logic, you’d end up writing unique lines of code and creating unique displays for each piece of equipment. Fortunately, most industrial cases are not like this – equipment tends to have high degrees of similarity, or there are common problems like monitoring paired signals or operational envelopes.

   To get the best value out of PI AF, focus on scalable templates. In 10 hours, you have a basic template. Within 100 hours, you have a good template and display for one asset. By 200 hours, you’re now monitoring twenty assets, or fifty, and your time to value is shrinking rapidly.

2. Integration and extensibility:

   PI AF has an immensely rich feature set, with loads of gems hidden under the bonnet. A few examples:

   i. PI AF Notifications don’t just send emails – they can ping REST endpoints to trigger requests and actions by other systems.

   ii. PI AF Tables can grab data on-demand from all kinds of external SQL sources.

   iii. PI Universal File Loading (UFL) can gather in data from remote systems.

   iv. PI AF Event Frames capture events and comments, creating event-driven views of operations.

   v. PI Vision and Extensibility enable all kinds of standardisable, customisable visualisations. Don’t build 50 displays when one will do.

   vi. Unit of Measure Databases allow you to standardise how UOMs are handled, meaning you can templatise even where equipment use different UOMs.

   vii. The PI System Access pack adds all kinds of integration options. The AFSDK library is vast and flexible, and can be used to extend the PI System further.

I once worked with a truly excellent technical leader – he asked solid questions, could troubleshoot any issue, and create incredible features. He was a leader whenever he played in this space – but when managing a project or stakeholders, the project and team suffered.

Many games feature the ‘Class Triangle’ – if you select the Archer, give them a bow and a cowl, not a wizards’ staff or lots of heavy armour! If you can, field a team of all 3, so that you don’t have glaring weaknesses to compensate for.

Users, developers and managers all have strengths, weaknesses and special skills – use them wisely. Lean into the energy and strengths your team and business have… but figure out where you’re weak too.

Ensure you’ve got your critical classes covered:

• The Ranger has the route planned out, using their Project Management skills to organise the party, keeping an eye on the horizon to decide the party’s general direction, assign tasks and coordinate actions.
• The Knight has fought all kinds of dragons and completed all manner of quests, and wields deep SME Knowledge like a weapon to ensure victory for the party.
• The Mage brings deep and arcane Product Knowledge to each technical challenge, casting arcane runes to create magical solutions to niche problems.
• The Paladin swears by a strict code of Quality management, wielding this as a shield to protect the party during their quests.
• The Bard captures the lore of your team and ensures the world hears the song of their epic victories, ensuring robust stakeholder engagement from beginning to end. They know the lore of the organisation and share this with the team at crucial moments.

If any one of these is missing, you’re already in trouble. These could be two people or five, but they need to be present, engaged and working together.

Projects which actively combine talent, interest and energy deliver outstanding products.

At first, you’re tempted to go straight for the final boss – the business only has enough time or money to do this once, so why not play the entire game in one marathon burst?

So you get up at 6am, start playing, and hope to finish the final boss by dinner-time. Almost immediately, you get pulled aside by tutorials, side-quests and repeating sections of the game that weren’t as simple as they looked. By lunchtime, you’ve given up on achieving your goal in the original timeframe, and you’re now tempted to quit the game entirely – game over.

A quest is always (as its name suggests) a journey. For large projects, the solution and team grow and change between inception and delivery – build up your experience, collect your key items and learn how the game works. Build up your strength before taking on the final boss.

If you haven’t done anything like it before on a small scale, choose your quest wisely. Start with what you are certain you can achieve – one piece of equipment to monitor, one asset, or a small enough scope that you can contain it. It’s easier to go back to the business with a nugget of concrete value and ask for more funding than it is to return mid-project with no proven value saying it will require more time and money.

Deliver one small, successful project. Once it’s successful, build a larger one around it.