Bad data holding back forecourt digital transformation

Without good quality asset data, fuel retailers won’t be able to drive the next wave of digital transformation and maintenance automation within their business. Tom Caldwell, CTO at Techniche, examines the key challenges retailers face as forecourts become more complex places to maintain.

Spreadsheets are dead

When fuel retailers embark on their journey to make maintenance smarter, a common stumbling block is often the bad state of asset data. Made worse by the fact that many are still using spreadsheets to try and manage their asset data, which simply doesn’t work. Many don’t have a complete view of what assets they have and until rectified, the introduction of new technologies including AI (Artificial Intelligence), machine learning and IoT (Internet of Things) is impossible.

This quest for better asset data is a persistent problem. But it’s compounded today by the business drive to introduce new technologies to achieve more automation, increased efficiencies and better operational insight across the forecourt, as new modes of transport and consumer behaviour begin to take hold. And when technology vendors come along offering innovation in the shape of AI, retailers will need to be able to offer up complete, clean and usable asset data in the cloud, to feed into a raft of exciting new applications.

Data holds the key

Creating and maintaining consistent asset data with effective hierarchies is one of the first and biggest challenges retailers face in solving the problem. Not creating and maintaining quality asset data has a negative impact on the outcome of reporting and analytics, which in turn results in low levels of intelligence on maintenance operations.

As a result, many retailers cannot accurately identify the asset fault trends taking place across their estate as they don’t have a single source of truth. Asset data is either not recorded, recorded inconsistently using generic terms, is incomplete with fields such as ‘location’ or ‘condition’ missing, recorded manually on paper or is held in different siloed systems.

Most modern assets can connect to a computer network but older or analogue assets can’t as they’re not ‘smart’. These assets still need to be integrated into any up-to-date digital inventory and many organisations resort to time-consuming physical visits to gather information, which is inefficient and results in patchy data at best.

What does a good asset register look like?

Variables such as how to categorise assets and how granular inventories should be, often create a lot of confusion around what a good asset register should look like. Plus, fuel and convenience retailers sort their asset data differently, applying their own interpretation of categories.

But to achieve reporting of any value, specific categorisation of assets is needed too. A balance is required between usability and getting good data back, so the categorisation and hierarchies need to be right. In the past, some global fuel retailers might have had over 3,000 different repair codes to log different faults. But asset data this detailed isn’t pragmatic – staff using the system can easily pick the wrong option or simply opt for the first and most convenient item in a dropdown menu.

Populating a detailed asset data template is much easier than starting data collection from scratch. In addition to accelerating the process, it helps include details that might otherwise have been overlooked, as the many different asset categories and hierarchies have already been created. Urgent has over 20 years’ experience of understanding asset performance and builds this experience into its asset templates.

Once an asset data hierarchy is established, maintaining the inventory has to become part of day-to-day operations for it to stand a chance of surviving long-term. The use of mobile apps within the asset maintenance workflow are a good way of encouraging staff or contractors to update data when out in the field. Network discovery tools can also be used to find assets on an IT network which might have been ‘hidden’ or forgotten about, or identify rogue items which perhaps shouldn’t be there.

Benchmarking asset performance

Once a meaningful asset register has been achieved, the next stage is to examine it across multiple sites, enabling something which has not been possible before – the benchmarking of different assets against each other. Comparing the performance and lifespans of different brands and models across an estate means budget forecasting and planning for replacements is dramatically improved, as trends can be monitored. The impact of environmental variables such as altitude or climate, for example, can also be measured to help determine which assets perform best in different locations. As the forecourt becomes more complex and reducing costs becomes even more pressing, this proactive management of assets is where the focus needs to be.

What does the future hold?

New approaches to gathering asset data, which make use of AI and machine learning, will soon be automating parts of the asset inventory management workflow. AI systems will be able to recognise assets from photographs and determine what they are – a far quicker way of creating an up-to-date asset register and reporting faults. This also eliminates human error when site staff record or input data.

AI will also help to shift maintenance strategies from reactive and preventative maintenance to predictive maintenance. For example, taking data feeds from IoT connected assets, AI can spot trends and anomalies long before something breaks down.

Consider air-conditioning units, which consume more power when the filters become blocked. AI and machine learning will be able to tell at what point it is best to replace or clean the filters – the balance between increased electricity cost and the cost to repair or intervene. AI can monitor the unit far more closely than a human ever could.

Data transformation

The ongoing digitalisation of fuel retail maintenance rests on having a solid foundation of asset data. Something which spreadsheets simply can’t deliver. Working with fuel retail asset data for over 20 years has shown that once a proper, workable data structure is in place and repair jobs are logged against it, the effectiveness of maintenance management is transformed. This actionable insight translates into better control of repair schedules and budgets, which in turn means reduced downtimes and a superior customer experience.

With the forecourt landscape set to change dramatically over the next few years, now is the time to examine how best to improve asset data before things become a lot more complex.

By Tom Caldwell, CTO at Techniche, a technology company which has been at the forefront of maintaining the critical assets of global fuel retailers for over 20 years.