Artificial Intelligence (A.I.) is, along with the Industrial Internet of Things (IIoT), the base of the ongoing wave of technological innovation denominated 4th Industrial Revolution or Manufacturing 4.0. Since its inception, AI has been surrounded by hype and misconceptions. This article aims to clarify what A.I. is in reality, its evolution, technological maturity and adoption, as well as its present and potential impact.
A.I. DEFINITION
Intelligence and Artificial Intelligence
Intelligence can be roughly defined as learning, understanding, and using learned knowledge to achieve goals. Thus, artificial intelligence can be initially and simply defined as intelligence exhibited by machines.
Artificial Intelligence (AI) is a branch of computer science that builds programs and machines that can creatively solve problems and are capable of performing complex tasks. They do so by automatically combing and analyzing large amounts of observational data, and applying algorithms designed to find patterns.
AI and Machine Learning
Besides the proven capability of processing enormous data sets, AI is characterized by making it possible for machines to learn from experience and adjust to new inputs. Machine Learning is achieved by applying iteratively complex mathematical calculations which are addressed to recognize patterns. AI adapts through progressive learning algorithms and finds structure and regularities in data, so that the algorithms acquire a skill: The algorithm becomes a classifier or a predictor.
To do so, iteration on large sets of data is essential, because when new data is provided, the AI System applies what it has already learned from previous computations, iterates the computation on new data and produces more reliable results and decisions each time.
AI System learns only from the data provided. This means that any inaccuracies and obsolescence in the data set will be reflected in the results, and therefore any additional layers of prediction or analysis need to be added separately.
As a result, most of today’s AI systems are trained to be limited to do a clearly defined task. These AI Systems are highly specialized: The AI System that detects insurance fraud, for example, cannot cannot detect tax fraud.
A.I. EVOLUTION
The AI concept has been surrounded by hype since it was created, along with the first neuronal mathematical models, in the late 1950s. Despite significant scientific advances, it was only in the 1990s that the hype began to be justified. In the 1990s, Machine Learning became really functional and was quickly refined, leading to its massive adoption and use by the manufacturing sector.
Machine Learning Advancement Through the Integration with other Technologies
The improvement of AI accelerated during 2000s-2010s, further developing Machine Learning. However, it is only after 2010 that AI learning capacity has accelerated exponentially thanks to the application of BigData in the Machine learning process, since the more data is analysed, the more the AI system learns. Distributed Cloud Computing, provided by Digital Platforms offers the massive computing power necessary to train algorithms and more complex multi-layered models.
Although impressive, AI development based on Machine Learning remains limited to solving specific problems and cannot tackle general problems in the same way as humans can.
Machine Learning Refinement Techniques
Learning is further improved applying innovative Data Training methods for a system to learn the relationships of a set of given inputs to a given output, this is used for example to recognize objects within an image or detect objects and describe their content. There is also the field of Reinforcement Learning to train AI System to learn from trial and error, by giving the AI System virtual rewards and punishments through a scoring system.
Currently, there exists justified excitement over the exponential refinement and growth capability of Machine Learning by applying these new learning techniques. Additionally, and more importantly, there is also the emergence of a more advanced field in AI, known as Deep Learning.
A.I. AND THE NEW DEEP LEARNING
Deep Learning is the most advanced level of machine learning, it is based on the addition of Neural Networks to the AI System, that simulate interconnected biological neurons and that model the way that neurons interact in a brain.
This new neural-network provided Systems accept and apply innovative learning techniques, which allows AI to classify, recognise, detect and describe the surrounding environment. In short, AI Systems have been provided with something close to the capacity of understanding.
This Deep Learning allows State-of-the-art AI to achieve incredible accuracy which was previously impossible. For example, our interactions with Google Search or Alexa are based on Deep Learning, and they become more accurate the more we use them.
A.I. GENERALISED USE AND EFFECTS
AI is allowing for an automating analysis which was not possible before, with impacts across all business and administration departments. It also allows for sets of complex tasks and functions to be conducted in fully autonomous or semi-autonomously way by AI enabled automation and robotics machines. Additionally, AI is at the core of unmanned vehicles and drones.
By 2030, Artificial Intelligence is estimated to add $15.7 trillion to the global GDP, with $6.6 trillion projected to come from increased productivity and $9.1 trillion from consumption (1). Businesses are aware of it, and by 2017 an estimated 20% of firms worldwide were using AI at some level and another 40% were experimenting with how to deploy it (2). The diagram below illustrates AIs broad range of technological maturity and and adoption by the transport and logistics sector (3):
CETMO Analysis 2021, adapted from McKinsey & Company (3)
For business, the largest impacts are, and will be, in the logistics and supply chain sectors followed by the manufacturing sector not only in production and operations but also in marketing and sales through personalization and niche adjustment.
For administrations AI’s enhances and deepens human planning and decision-making, which has an undeniable potential to optimally leverage scarce resources and improve their social effects.
As in many other of the Industrial Revolution 4.0 there is an ongoing and intense debate among experts regarding whether AI will consolidate existing business competitive advantages among countries and corporations, or will level the playfield advantages.
In any case, the general conclusion is that in order to avoid being left behind in competitive terms, corporations, operators, administrations and regulators should explore and accumulate knowledge about this technology, albeit on a small scale, in order to obtain direct experience at low cost.
References:
- Brookings Institute. Africa Growth Initiative. Travaly and Muvunyi, the future is intelligent: Harnessing the potential of artificial intelligence in Africa, 2020. link
- Bughin, Chul, McCarthy, How to make A.I. work for your Business, 2017
- McKinsey & Company. Ashutosh, Hastings, Murnane, Neuhaus, Automation in Logistics: Big Opportunity, Bigger Uncertainty, 2019. link
ARTIFICIAL INTELLIGENCE AND BIG DATA EFFECTS ON TRANSPORT
Efficient access to AI is done mostly through Digital Platforms based on Cloud Computing (CC). Also, much of the work done by AI is combing, analyzing and learning about the data provided by BigData.
BigData is a mature and widely adopted technology. Therefore, Big Data can be considered as part of the necessary infrastructure for the operation and deployment of Artificial Intelligence and also of other new technologies such as Robotic Automation, Internet-of-Things, Autonomous Vehicles and BlockChain.
The fact that this infrastructure is offered at low cost and even free of charge through digital platforms is the democratizing element that drives the technological advancement of all these technologies, as well as their adoption and the extent of current and potential effects.
A.I. Effects on Freight Transport
Decision Making Support: Artificial Intelligence, particularly if applied to Big Data Analytics enables advanced business intelligence tools, Advanced predictive modelling and generation of indicators and dashboards among others, showing an accurate picture of the company and situation, thus allowing for decisions that are better informed and less risky.
Semi-Autonomous Security and Supervision of infrastructures and terminals increasing the control capabilities with sensors of temperature, humidity, and other environmental control metrics, thus allowing for the control of large spaces at a lower cost.
Route and Price Setting Optimisation: with greater adjustment to costs and commercial strategy for each of the network routes, that allow for the optimisation of each case and expanding the possibilities of yield management.
A.I. Effects on Passenger Transport
Decision Making Support: in the same way that it affects Freight Transport and other industries, Artificial Intelligence will enable the use of advanced and more reliable predictive models, making planning easier.
In the most forward-looking scenario, AI based modeling and other AI advanced tools may even replace pre-planning with an instant, continuous, and automatic adjustment of the service, responding to evolving needs, saving time, and increasing quality.
Safety Improvement and Maintenance Cost Reduction: AI allows for predictive maintenance of infrastructure and rolling stock, like Infrastructure MRO (Maintenance, Repair & Overhaul), rolling stock MRO, as well as energy consumption optimisation.
Artificial Intelligence and Passengers: AI, along with facial recognition and other biometric applications, will optimise processes that require identity checks, such as the check-in of luggage, documentation, and border control, and should allow for faster, less error-prone traveler experience and administrative procedures.
BIG DATA EFFECTS ON TRANSPORT PROCESSES
BigData Effects on Freight Transport
Improved Planning: searching, retrieving and processing large volumes of data in real time (such as costs, constraints, limitations, asset location, etc.) helps to improve operations forecasts and support tools (such as freight market, fuel purchase, weather forecast, travel time, etc.).
More Accurate Asset Lifetime Analysis: lifetime analysis and preventive maintenance of machinery and other movable assets is more accurate reducing fix costs and improving the Balance Sheet.
Network optimisation: data analysis leads to optimisation of routes and the reduction in general of operating costs and future investments.
Operations Fine-Tuning: information support and measurement of operations allow for realistic criteria to be established, adjusted to the real circumstances, and the improvement of the design and control of operations.
New Skills Required: as there is a need to incorporate new technical profiles in fields related to mathematical computing, Data Science and Operations Research.
BigData Effects on Passenger Transport
Improved Planning and Fleet management: similar to how Freight Transportation Big Data Analytics allows for the improvement of operations, forecasts, and support tools. Additionally, it permits the optimisation of decision making by providing more and better indicators, and to develop new products and services from a new perspective that is more adapted to the real needs of travelers. This allows for the optimisation vehicles, routes, and costs (instantly, or with future simulations).
More Accurate Asset Lifetime Analysis: same as Freight Transport.
Trends, Micro-Segmentation and Improved Customer Service: Big Data Analytics permits forecasting of megatrends, changes in customer behavior and micro-segmentation. It also provides integrated, customised information to the customer while purchasing and travelling..
New Skills Required: similar to Freight Transport.
Sources: CETMO and “Impacte de les KETs en la digitalització dels diferents àmbits del transport”, CENIT-CINESI – December 2020
With the support of:
With the support of:
Subscribe to our newsletter
Stay updated about all news and events