The overall theme of our recent webinar on AI data transformation focused on creating a safer, more efficient, and sustainable supply chain and logistics. At Anaeko, we work with transport and logistics companies to help them analyse and share data effectively.

From a safety perspective, the goal is to reduce incidents and accidents by integrating AI technology into trucks and other technologies. This AI can detect hazards and raise awareness, support assisted driving, and eventually lead to autonomous driving. Compliance is another key aspect, where understanding and meeting training needs are essential to protect drivers and the environment.

Efficiency is achieved through route optimisation, ensuring that logistics operations are as productive as possible. In supply chain management, predictive maintenance is a significant focus. By analysing past data, AI can help predict and prevent future maintenance issues, increasing the utilisation of vehicles and equipment across the UK.

Currently, 40% of freight journeys are made with empty trucks. Optimising routes and planning can greatly improve vehicle utilisation and better manage drivers, who are a scarce commodity. This efficiency drive is crucial for the industry.

From a sustainability standpoint, analysis should ultimately benefit the environment. Reducing carbon emissions and optimising fuel usage not only cut costs but also help create a greener and cleaner environment.

The Journey to AI

AI is a journey built on a solid foundation of data. The blockchain ID provides an immutable record, ensuring trust and preventing tampering. This reliable data offers detailed, accurate tracking information, which is essential for AI. While humans can account for nuances in imperfect data, AI requires perfect data to function effectively.

When implementing AI within an organisation, there are two key components. First, AI needs to be integrated into existing systems and platforms. Second, to drive specific AI processes, it is crucial to raise awareness within the organisation and among staff about what AI is and isn’t. Experimentation is necessary to prove the value of AI. Often, we implement pilot projects where measurable impacts are demonstrated before scaling operations across fleets, regions, and supply chain functions.

There's a maturity in the use of data and analytics, and typically, AI adoption increases along that maturity curve. Transport and logistics firms are transforming their use of data from historical reporting to more advanced analytics.

Initially, organisations focus on descriptive analytics, which report on what happened. For example, performance indicators might be tracked weekly, monthly, quarterly, or annually to assess and improve safety targets. This involves looking back at historical data to identify trends and areas for improvement.

As organisations progress, they move towards diagnostic analytics, which aim to understand why something happened. This stage often involves machine learning and AI to analyse much more data than a human could manage. By surfacing insights, these diagnostics help organisations make better decisions.

The next stage is predictive analytics, where organisations look at trends to forecast future outcomes. For instance, correlating checks performed and defects found during predictive maintenance can help identify patterns and predict future issues.

Finally, organisations move towards prescriptive analytics, which recommend solutions based on the data. This is particularly relevant for sustainability in the supply chain. For example, EV replacement strategies can benefit from economic life cycle analysis of vehicles. By understanding the right time to replace vehicles and which vehicles to choose, organisations can make data-driven decisions. This analysis can be enhanced with data from blockchain technology, which provides detailed information about vehicle efficiency and other metrics.

AI Requires Information Architecture

We are a professional services company, and we say that AI requires Information Architecture (IA). This architecture connects to the internal systems within your business, facilitates partner data sharing, and integrates with vehicles and staffing to ensure you get trusted data.

This data needs to be findable, usable, and machine-readable for AI. It also needs to be understandable. Within this architecture, there are standards that can be adopted to help share information across the supply chain and ecosystem. In the absence of these standards, metadata becomes essential.

Data Optimisation for Autonomous Driving

Anaeko worked with autonomous driving, where there are five levels of automation, ranging from no automation (manual cars) to full automation. Currently, most of us are familiar with driver assistance and partial automation. These technologies make roads safer by detecting hazards between vehicles and staff, promoting sustainability through better control of harsh braking and acceleration, and improving accessibility. They also drive economic growth through increased efficiency and create new jobs.

For example, we worked with a vehicle manufacturer to pilot a project that analysed data from their sensor-equipped vehicles. The goal was to ensure these vehicles were safe, efficient, and sustainable. In Big Data, there are five key characteristics to consider:

1. Velocity: Processing sensor and video data quickly.
2. Variety: Handling different types of data from various sensors and cameras.
3. Veracity: Ensuring the quality, integrity, and accuracy of data despite potential sensor failures or weather conditions.
4. Volume: Managing large amounts of data, with more data generally being better for analysis.
5. Value: Extracting useful insights from the data to inform decision-making.

In this project, data scientists analysed scenarios, such as right turns at T-junctions, using data engineering to ensure privacy and security by redacting driver and vehicle identification details. Managing the cost of large data volumes and accelerating time-to-analysis were crucial.

In practice, this involved identifying and classifying potholes, signs, and other vehicles using computer vision and sensors. The trained AI models then applied these classifications to improve vehicle performance. Companies like Tesla and Volvo are implementing similar techniques in their autonomous driving systems, using hazard detection to create cleaner and safer transportation experiences.

With great power in AI comes great responsibility. It's essential to ensure that AI-driven decisions adhere to ethical principles, laws, regulations, and maintain integrity. As AI's popularity increases, so does its power consumption, which necessitates sustainable practices. Cloud providers like Amazon Web Services and Microsoft are advancing greener energy consumption to mitigate this impact. While there may be concerns about AI taking jobs, in reality, it's about augmenting and cooperating with humans.

To deliver impactful AI solutions in business, you need an architecture built on trusted data, where blockchain plays a crucial role. This involves transitioning from historical reporting to prescriptive analytics that make and recommend decisions. Expert skills in AI, data engineering, and data science are vital for this sophisticated work. Transport and logistics firms should focus on their core business, offloading the heavy lifting to technology specialists.

To achieve measurable business impacts, initiatives must be designed to deliver specific benefits. Kieran's single pane of glass provides a comprehensive view of efficiencies gained, hours saved, and frictionless borders, contributing to a cleaner, more efficient, and safer transport and logistics supply chain.

Book a one-on-one consultation with the Anaeko to team to start your journey to AI:

Join our next webinar: