Lord Kelvin, the flying chimney and artificial intelligence
How international standards support innovation in the transport industry
Lord Kelvin, the founding President of the IEC, recounted an amusing anecdote about his father, in a speech to the University of Glasgow in 1904. According to Kelvin, one day when his father was still a young man, he was astonished to see what appeared to be a black chimney moving swiftly along the River Clyde. It turned out to be a paddle steamer, which was no doubt amusing to the audience as for them steamships were a common sight.
Lord Kelvin used the story to illustrate how technological advancements, such as steam power, quickly became so commonplace that they no longer elicited excitement or curiosity. This also holds true in today’s world. Artificial intelligence, for example, has become so deeply ingrained in our daily lives that we take it for granted.
The transport sector is a case in point. Business operators are leveraging big data and technologies such as machine learning and neural networks to optimize routes, reduce costs and enhance safety, while improving customer experience. This article discusses some of the ways that AI is being used in the transport industry. It also looks at how the IEC and ISO committee on artificial intelligence, known as SC 42, is supporting the innovation.
The examples that follow are based loosely on the use cases presented in ISO/IEC TR 24030. The technical report published in 2021, includes over 130 AI use cases, from 24 application domains, including transport. SC 42 is currently revising the publication with new and updated use cases.
Autonomous vehicles
Autonomous vehicles are one of the most significant applications of AI in transport. These vehicles use a combination of sensors, cameras and machine learning (ML) algorithms to navigate roads and make decisions without human intervention. ML algorithms analyze vast amounts of data collected by sensors and cameras to identify patterns and make decisions based on those patterns.
For example, ML algorithms can recognize traffic patterns and adjust vehicle speeds to optimize fuel consumption and reduce emissions. Additionally, autonomous vehicles can use neural networks to process large amounts of visual data in real-time, enabling them to detect and avoid obstacles, pedestrians and other vehicles.
Traffic prediction and management
AI can be used to predict traffic patterns and optimize traffic flow. Traffic prediction algorithms analyze vast amounts of data collected by sensors installed on roads, public transit systems and other transport infrastructure. This data is then used to predict traffic patterns and identify potential bottlenecks.
ML algorithms can analyze historical data to identify patterns and predict future traffic volumes. This information can be used to optimize traffic flow by adjusting traffic signals, rerouting vehicles, and managing public transit schedules.
Predictive maintenance
AI can be used to perform predictive maintenance by analyzing data collected by sensors installed on vehicles and infrastructure. ML algorithms can analyze this data to identify patterns that indicate when maintenance is required. This can help reduce downtime and prevent breakdowns, improving both safety and reliability.
For instance, machine learning algorithms can identify the patterns of vibration and temperature that indicate that a vehicle’s engine is likely to fail. This information can be used to schedule maintenance before a failure occurs, reducing costs and improving safety.
Supply chain management
AI can optimize logistics operations, including route planning, delivery scheduling and inventory management. Machine learning algorithms can analyze vast amounts of data collected by sensors and GPS trackers to optimize routes, schedules and inventory levels.
For example, machine learning algorithms can analyze traffic patterns, weather conditions and other data to identify the most efficient route for a delivery truck. Additionally, neural networks can analyze data from multiple sources to provide real-time information about the availability of inventory and the status of deliveries.
Customer service
AI-powered chatbots can help answer customer queries and provide real-time information on transport schedules and delays. Chatbots can analyze large amounts of data to provide customers with the most relevant information and answer questions about transport services.
For example, chatbots can use natural language processing algorithms to understand customer queries and provide real-time information on public transit schedules, fares and delays. Additionally, chatbots can analyze customer feedback to identify areas for improvement in transport services.
Energy efficiency
AI can optimize energy consumption in transport, such as route optimization to reduce fuel consumption and emissions and smart charging systems for electric vehicles. Machine learning algorithms can analyze data such as traffic patterns, weather conditions, and vehicle performance to optimize routes and reduce fuel consumption.
For example, machine learning algorithms can analyze data collected by electric vehicle charging stations to predict demand and optimize the use of charging stations. Additionally, neural networks can analyze data from multiple sources to predict weather conditions and adjust energy consumption accordingly.
International standards
International standards are crucial for enabling current and upcoming innovations in various industries, including the transport sector. SC 42 aims to cater to the diverse needs of the entire AI ecosystem. Their publications not only cover technical aspects but also non-technical requirements, such as regulatory and policy issues, business considerations, ethical and societal concerns, and application domain needs.
Drawing on the expertise of a broad range of stakeholders and IEC and ISO domain experts, SC 42 develops international standards that address the evolving challenges and requirements of diverse technologies and applications. Recent ISO/IEC publications cover topics such as AI-related risk management, bias in AI, big data analytics, computational approaches, ethics, governance and ML classification models.
SC 42 also organizes bi-annual ISO/IEC AI workshops that target a diverse community of stakeholders interested in AI and the committee’s work. These workshops serve to promote the committee’s work, scout emerging trends, identify requirements, offer insights, and provide opportunities for future standardization work. The workshops also aim to attract new stakeholders and foster discussion on AI-related issues. The next workshop is scheduled for June.
