The retail industry is taking on a new challenge in raising the bar on customer experience. The convenience store (c-store) experience is beginning radically change with some of these new ideas. The traditional gas station, for example, is being challenged as we introduce more electric cars into the market. Even in truck stops, we may begin seeing self-driving trucks pull up that need to be fueled. Rest stops may evolve into high-end automated self-service mini-malls. The digital transformation (DX) journey for retail will include the use of artificial intelligence (AI) combined with IoT devices to drive automation and augmentation for the next generation of products and services in the c-stores of the future.

We are already seeing new concepts emerge like Shell Oil’s new “Shell Select” store, which offers a consumer-focused retail store experience with locally inspired fresh foods and beverages in a relaxing and friendly environment.

The “Good Food Market” in Ireland won the c-store of the year by challenging the norm for urban convenience retailing. It operates as an upscale, urban grocery store that also has a changing food offer throughout the day, all-day dining for in-store eating, gourmet take-home meals and world cuisine-themed evenings. Change is occurring across the globe, with all brands planning out their stores of the future.

Digital transformation, including AI and IoT, will be a core technology evolution in driving the successful change in customer experience.

At RSA 2020 this year, I co-hosted a talk with Darren Bennett, the deputy IT director and CISO for the city of San Diego. We discussed how light industry IoT is evolving smart buildings, smart campuses and smart cities into the “city as an IoT platform.” A platform of platforms with new IoT devices being added every day, along with their SaaS back-end clouds, to drive new capabilities and risk.

At Forecourtech ’20 in October, I will be describing the dangers of light industry IoT in the oil retail industry to how to leverage AI successfully. AI is a critical part of this retail digital transformation, especially as it applies to the IoT devices, and there will be new challenges in regard to the area of IT operations management (ITOM).

A new emerging technology named AIOps unifies operational teams (facilities, NetOps, SecOps) when managing DX. Whether it is a gas station, a mini-mall, a campus of buildings or an entire smart city, there are many “things” out there attached to buildings and connected to networks that have software in them to make them intelligent. I call this light industry IoT. It can include smart air conditioners, CCTV cameras, coffee machines, sensors, gas pumps, signs and even lights — all those “things” that facility teams install that have software and are commonly overlooked by the IT operations teams.

However, these devices could become the primary entry point of your next cybersecurity breach. In 2019, the WaWa brand of c-stores had its point of sale (POS) system breached, resulting in over 30 million compromised credit cards. The famous Target breach was from hackers coming in through the network-connected air conditioning system and then attacking the POS systems. AI is both a defensive and offensive strategy in digital transformation.

As we look closer into the expanding risk footprint of new DX technologies, especially in smart cities and oil and gas retail, the addition of new IoT smart devices helps enrich the focus on the customer experience. With the change, IoT brings along many new risks around these digital technologies, including the IoT devices themselves, cloud computing, customer applications, data privacy, partner applications and new sensors. These new risks will require process changes, modified priorities and even staff modification to manage the risks associated with new IoT technologies.

In parallel with driving DX change, operations teams must begin a transformation of their people skills, operational processes and tools. Collectively, DX requires improved operational monitoring capabilities to automate and create greater efficiency in managing this new risk footprint. AIOps (the use of AI in IT operations) starts with the ingest of big data (clean data). The data could be at rest and be used for historical analysis, or the data could be in motion (streaming) for real-time analysis. The algorithms then look for prescriptive responses that can turn into actions. AIOps is not “yet another technology tool.” It comes with automation algorithms for predictive analytics, the correlation of alerts and the remediation of issues.

AIOps is not a perfect tool. Just like humans, AIOps is prone to errors. Unlike human errors, AIOps is prone to various types of digital errors found in error-prone data sets and biased training of machine learning models. Furthermore, there is no audit trail of AIOps if it leverages neural nets or deep learning algorithms. These types of machine learning models are “black boxes” and cannot show how their recommendations were reached.

Today, operations staff, whether it is facilities, NetOps or SecOps, sift through millions of alerts and try to predict, correlate and remediate an actionable response. With AIOps automation, the remediation algorithm must understand causal inferences. Meaning if A causes B, then AIOps needs to predict A to prevent B. The AIOps algorithms from prediction, correlation and remediation need to choreograph together and augment the human IT operations staff into becoming Jedi warrior IT analysts.

As far as light industry IoT goes, the operational team must begin to take steps and look at the array of smart devices related to facilities in their organization and then plan a fully automated AIOps-driven life-cycle maintenance and monitoring strategy for those “smart” and connected assets. As a unified strategy, the SecOps team needs to design best practices for cyber hygiene (such as NIST). In theory, this should all be possible through AIOps automation without requiring additional staff.