Introduction
Technologies have been problem solvers for businesses in the past; whether it is retail, banking, insurance, healthcare or even sports. Some of these solutions have changed the way businesses are being run through reduction in operating cost, improving efficiency, and increased efficiency. One of the upcoming areas where latest technologies have been successfully implemented is transport industry which is plagued with issues related to traffic congestion, unexpected delay and routing problems leading to monetary loss in organizations. Transport Industry has been a major contributor to the movement of people and goods across various geographical regions. It plays a significant role in supply chain management system where goods are transferred from one place to another. The industry plays a key role in the movement of goods to the right place at the right time in a logistics chain. In order to reap the complete benefit from a business investment, technologies like Machine Learning, Artificial intelligence, internet of Things among others have been used by governments and organizations.
Artificial Intelligence (AI)
Artificial intelligence (AI) is a broad area of Computer Science that makes machines function like a human brain. AI is also defined as the ability of a machine to perform cognitive functions of a human at ease. The phrase AI was initially coined in the year 1956 by John McCarthy, a computer scientist. This six-decade old concept has gained recent buzz due to the availability of large volumes of data generated through various devices, and availability of efficient hardware, software and network infrastructure. The advent of AI has enabled process automation leading to innovative business solutions.
AI provides reliable and cost-effective solutions while addressing uncertainty in the decision-making process. The ability of advanced algorithms to handle complex data has facilitated faster decision-making in businesses due to process automation. With the growing concern related to environment, AI has become a solution provider to resolve climate change and water issues by transforming the traditional sectors and systems. These capabilities have helped governments build sustainable cities that would help protect biodiversity and wellbeing of humans. The United States (US) and China currently dominate the world of AI.
A PwC report estimates that AI will contribute $15.7 trillion to the world economy by 2030 – more than the combined current output of China and India. In the US, the academic system has generated and incubated research related to AI; whereas in China, funding and technology is provided by the government to utilize the potential of AI. China plans to invest at least $7 billion till the year 2030. Canada and the United Kingdom have ramped up investment in technology by announcing deals to fund private and public AI ventures. Canada had made $125 million commitment in 2017 itself for AI research. The French Government is investing $1.8 billion in AI research until 2022. The country plans to extract data from private companies to be publicly available for research. Russia spends an estimated $12.5 million annually on AI predominantly in military. Among developing countries, India is uniquely poised to be a leader in AI in the next few years. This is due to its strength in technology, favorable demographics and structural advantages due to the availability of data generated through digital transactions. The Indian government set aside $477 million for Digital India project to enhance focus on AI, IoT, big data among other technologies. One of the significant use cases being traffic and crowd management.
AI and Transport
Most of the large cities across the globe face issues related to transport, traffic and logistics. This is due to the fast-growing human population and also due to the increase in the number of vehicles on the road. In order to efficiently create and manage a sustainable transport system, technology could be of immense support. With urban areas struggling with traffic congestion, AI solutions have emerged in accessing real-time information from vehicles for traffic management, and utilizing mobility on demand in trip planning through a single user interface. Safe integration of AI-based decision-making, traffic management, routing, transportation network services and other mobility optimization tools are other possibilities of efficient traffic management. AI is considered as one of the emerging technologies by World Economic Forum. AI methods that support transportation include Artificial Neural Networks (ANN), Genetic algorithms (GA), Simulated Annealing (SA), Fuzzy Logic Model (FLM) and Ant Colony Optimizer (ACO).
The objective of deploying these techniques in transport management is to relieve congestion, make travel time more reliable to commuters and improve the economics and productivity of the entire system.
Vehicles that are connected through technology improve driving efficiency through forecasting of traffic conditions on the road.
The research article addresses three perspectives:
- Assessment of accurate prediction and detection models aiming to forecast traffic volume, traffic conditions and incidents
- Public transportation as a sustainable mode of mobility by exploring various applications of AI
- Connected vehicles aiming to enhance productivity by reducing the number of accidents on highways.
Several studies have been conducted across the globe to overcome issues related to the transport industry. The outcome of the research activities with the support of AI technologies around this industry has given hope for this significant area of development.
Intelligent Transportation
In the recent years, huge volumes of data are generated with the proliferation of multiple technology devices across sectors. This data has become valuable in the decision-making process of businesses, governments and societies. Transport industry being the life line of an urban set up cannot be left behind in data generation and usage. This sector plays a significant role in urban development because it impacts people, processes and profit. To enable data generation, automobile manufacturers have been pro-active in building devices that can be fitted into vehicles that are used for transporting people and goods. Data generated by these devices are monitored remotely by experts. Governments and businesses are capable of taking real-time decisions based on the data generated through using various applications. Various innovative applications related to transportation and technology are being built in the recent years. The application developers focus on a process-oriented systems approach with a clear goal embedded with a feedback mechanism to measure the outcome of the solutions related to transport industry.
Transport Management Systems (TMS) belong to the area of transportation management specifically concerning the transportation operations. The objective of these systems is to set up effective route planning, load optimization, improved flexibility and transparency using data. As per Gartner, this field is expected to grow at a fast pace. Transportation strategies of a city are linked to an information system for better administration that would focus on capturing, processing, transmitting and management of the data thus generated.
Since the past couple of decades, due to the emergence of smart technologies, various information systems for logistics, routing, mapping and planning are being developed. These systems have provided increased data processing capabilities to better plan the transportation process leading to Intelligent Transportation Systems (ITS). Data generated from the users and vehicles are used to build efficient ITS. Building of ITS into the transport systems has ensured increased performance due to information acquisition, exchange and integration across vehicles, city infrastructure and other related activities. It is observed that ITS supports the decision-making process for the city authorities and vehicle users.
The article focuses on Intelligent Transportation Systems that forms part of Transport Management Systems. A desk-based method is adopted up to collate AI techniques to resolve Transport industry issues towards building a sustainable transportation system. Benefits of the various subsystems of ITS are identified and discussed along with applications of AI which have impacted the Transport industry positively. Data is summarized from research papers, Government reports, Journal articles and reports from consulting agencies.
Literature survey
AI has caused significant disruptions in various industries like healthcare, retail, banking, insurance, entertainment, manufacturing and transportation. Several use cases of AI in transportation have been experimented and adopted justifying the fact that this market is on an upward surge. With the technology advancement related to AI, transport industry has transformed itself into embedding user friendly devices into vehicles. This has led to the building of ITS using the data generated from the devices.
AI in the current form has the ability to solve problems in real time transport thus managing design, operation, time schedule and administration of logistical systems and freight transport. Some of the other applications include travel demand analysis, transport organization, pedestrian and herd behavior analysis. AI techniques allow utilization of these applications for the entire transportation management – vehicle, driver, infrastructure and the way in which these components dynamically offer transport services. AI methods provide smart solutions in areas where it is hard to fully understand the complex relationships between the characteristics of the transportation systems.
AI can be hardware based (robots) or software related (Google Maps). Data-driven AI combines machine learning techniques with technologies used for searching and analyzing large quantities of data. AI helps detect market trends; identify risks; ease traffic congestion; reduce greenhouse gas and air pollutant emissions; design and manage transport; and analyses travel demand and pedestrian behavior. Data and AI driven applications and services are the major cornerstones to achieve the vision of delivering optimal mobility. In order to build an effective and efficient mobility ecosystem in a city, a holistic approach of mobility management is required. Connected vehicles send data in real-time thus generating immense amount of data. With transportation demands continuing to increase, data growth through devices also grows; thus, creating a need for smarter management of road traffic.
The study suggests an AI based solution approach with multiple real-time knowledge related expert systems in arterial traffic management. Two AI paradigms – support vector regression (SVR) and case-based reasoning (CBR) are used for the evaluation of large-scale networks and complex simulation models. The study evaluated the outcomes of the two prototypes by comparing the predictions of traffic conditions. In this study, an agent-based control system monitors traffic, road incidents and other transportation activities.
The article compares two integrated autonomous agents deployed for intelligent traffic management systems that perform decision support for real-time traffic management around Barcelona.
The study investigates the applicability of autonomous intelligent agents in urban traffic control (UTC). The systems proposed by the study could design, implement, optimize and adjust UTC for dynamic environments. The utility of this model is suggested to be on several intelligent intersection of traffic signaling agents. These agents are capable of responding to real time traffic conditions and maintaining its stability and integrity. Technologies related to autonomous vehicles (AVs) have the potential to impact vehicle safety and travel behavior. They ensure reduction in travel time and increased fuel efficiency. Currently, these technologies have become disruptive in bringing immense benefits to the transportation system. However, challenges related to adoption by larger group of people and prohibitive costs of adoption remain. Regulations with respect to liability, security and data privacy are uncertain from the governments leading to lesser market penetration of autonomous vehicles. The study on the evaluation of the effectiveness of low-speed autonomous emergency braking system found that the vehicles fitted with this technology managed to reduce rear-end crashes by about 38 %. In the current scenario, major problems in transportation are congestion, safety, pollution and increased need for mobility. One of the potential solutions to tackle all these challenges could lie in autonomous vehicles. These vehicles collect data from their physical and digital surroundings through sensor technology and connectivity solutions.
Connected cars are capable of accessing the Internet through smart devices and are also capable of communicating with other cars and infrastructure. They draw real-time data from multiple sources supporting drivers through stressful operations during driving. These cars ensure safety and reliability. Pattern recognition is used with image processing for automatic incident detection and identifying cracks in pavements or bridge structures. Clustering technique is used for identifying specific classes of drivers based on driver behavior.
The article proposes new models, means, and forms of manufacturing of vehicles using technology. The study focuses on the process planning and the deployment of intelligent systems for manufacturing.
The study mentions that the initial demonstrations of technologies used in autonomous vehicles dates back to the year 1939. Most of the autonomous vehicles developed by the company Google rely on video cameras, radar sensors, laser range finders and maps developed by themselves.
Autonomous vehicles would impact the functioning of not only the individual companies but also the national and world economy.
Manufacturing and logistics generate huge amounts of data due to its networking capabilities with different stakeholders. With transport industry playing a major role in logistics it is only appropriate that the generated data is put to use through application of various technologies in the operations.
Future research is recommended through an agent-enabled supply chain optimization by the process of simulation.
The study ponders on the fact that whether driverless cars with AI have an adverse impact on humans. Extreme automation might lead to vulnerabilities in the machines. These integrated intelligent systems are vulnerable to systemic risks such as network collapse or hacking by external agencies. The study proposes Industry 5.0 which can democratize knowledge co-production from Big Data.
The research was undertaken to assist logistics managers, researchers and transportation planners to define and comprehend the basic views of logistics and its various applications and the relationship between logistics and transport. Transportation logistics is not only limited to the movement of goods across space and reducing time and costs along the supply chain. Its scope has expanded and it has become part of strategic management too. Hence, it is significant to integrate core business information systems with a set of modern analytical and AI tools to discover relevant knowledge from all sources. This helps in managing uncertainty and achieving competitive advantage.
The study incorporated various AI technologies to achieve four perspectives, namely – knowledge acquisition, service logistics, service automation and performance measurement. Transportation plays a larger role in building responsive logistics information system; hence, machine learning concepts support in identification of demand patterns and corresponding replenishment strategy.
The study signifies the systematic development of the process of the current logistics scenario. As we observe, intelligent systems have been playing a larger role in the logistics industry with transport industry as its backbone. The function of transportation has undergone a structural change making an impact on the evolution of transportation logistics. On the one hand, due to AI adoption, there might be huge reduction of accidents and fatalities on the roads; on the other hand, we can expect technological unemployment.
Frame-work
From the earlier studies, it is observed that the benefits of AI while building intelligent transportation systems has not been explored sufficiently. The current study explores the ITS applications in transport industry deployed in various countries.
Transport industry being the life line of an economy seems to be grappling with various operational issues across the globe. Issues related to transport industry have led to the slowing down of the progress of a city and in turn a country. TMS are a boon which are systems built to overcome transport issues using various technologies. TMS help businesses plan, execute and optimize the physical movement of goods.
Due to data availability and remote monitoring, TMS ensures timely delivery of goods leading to higher customer satisfaction. This has benefitted businesses through increased sales. TMS improves fleet performance and reduces supply chain expenditure with the use of appropriate tools like route optimization. Since data is collected remotely and keenly monitored, end-to-end understanding of deliveries, outcomes and returns are recorded leading to better transparency.
TMS uses technology to plan, execute and optimize the goods movement to help businesses thrive. These applications are used by manufacturers, distributors, retail businesses, and companies which are into logistic business. The major functions of TMS include route determination, outbound/ inbound logistic processes, route scheduling, 3PL vendor services, freight forwarders, service agents, transportation tracking and bulk processing of route scheduling & transportation planning. It is observed that the functions related to TMS are related to goods transportation. TMS integrates the multiple transport applications into a single package for better ease of use. TMS are made to be more intelligent using AI and Machine Learning to provide accurate predictions. Some of the technologies that are currently being used include: Internet of Things (IoT) devices and sensors, Digital assistants, delivery time prediction solutions, transportation planning solutions, Block chain among others. Intelligent Transportation Systems (ITS) have unfolded from TMS. A system which has the ability to take suitable decisions as per the given scenario using data generated from the devices installed in the vehicles are known as Intelligent Transportation Systems. Past studies have indicated that an integrated approach to ITS include the transport infrastructure and transport management.
ITS being a new type of TMS has been gradually replaced by automated control systems. They have evolved into forecasting of dangerous situations with the potential of being used as a decision[1]making tool with the use of huge volume of complex data. ITS has also impacted the efficient functioning of transport system through automated data collection in this dynamic environment.
A typical ITS requires input data from various devices and sensors. This data is monitored and processed remotely. The insights derived from the processed data is considered as a valuable input for governments and businesses to take decisions. This systems approach ensures continuous improvement in the performance through feedback mechanism. The input data is derived from the various devices fitted in the traffic management infrastructure, vehicles, and road infrastructure. Authorities monitor the data and ensure that timely data is disseminated to the commuters, drivers and pedestrians thus benefiting the stakeholders.
An Intelligent Transport System comprises of a set of sub-systems in the area of Public Transport, Traffic Information, Parking management, Traffic Management and Control, Safety Management and Emergency, and Pavement Management.
The study undertaken by Hamida et al, 2015 classifies the various applications of Intelligent Transport Systems into four main classes. They are
- infotainment and comfort;
- (ii) traffic management
- (iii) road safety; and
These applications collect data from vehicles to improve its utility thus ensuring driver safety and enhanced public transport facility.
ITS applications are generators of data which in turn assist in the decision-making process by the government authorities to manage public places in a better manner. Some of the applications are related to passenger comfort, improved driver experience, and efficient road management.
The ultimate beneficiary of Public Transportation System are the road users. The Intelligent Transportation Systems (ITS) framework for a sustainable public transportation system considers ICT technologies, automated transport system, traffic management center, and advanced traveler information system. Some of the applications which are built using ITS ensure traffic management, traffic signal control, vehicle navigation systems, smart parking management among others. ITS requires a network of technologies which operate together across the city infrastructure .The classification of problems of ITS as discussed by include performance monitoring, traffic management, improved transportation process, information support to participants of the movement, and transport infrastructure management.
Discussions
AI solutions for Intelligent Transportation
The contribution of AI to the field of transport industry has been immense and extensive. The solutions include autonomous vehicles, traffic management, optimized routing, and logistics thus providing safety of vehicles and drivers. ITS are built using the data generated from the devices installed in the vehicles through AI technologies. The current study focuses on four sub-systems related to transportation – namely, Intelligent Traffic Management System, Intelligent Public Transport System, Intelligent Safety Management System, and Intelligent Manufacturing & Logistics System.
AI accomplishments in transportation across the globe
As observed in the discussions so far, the capability of AI to solve problems related to transportation seems to be a natural fit. Yet, as is the case with AI in every other industry, the adoption of these applications varies across organizations and geographies. Based on the environmental and geographic factors, the applications could be both straight forward and complicated, distant and just-around-the-corner, definite or probable.
AI applications across organizations
United States seems to be fore-runner in these applications. This is probably due to lesser population and better road infrastructure as compared to developing countries like India. Start-ups which are innovative receive good amount of funding to develop prototypes in developed countries. Most of the solutions are experimented during long distance driving as compared to passenger vehicle segment.
Adoption of AI by transport corporations
As per AI is likely to have an increasingly drastic positive impact on city infrastructure by providing accurate predictive behavioral models of individual’s movements, their preferences and their goals. Though AI in transportation planning applications have become significant in the recent past, there is a concern of privacy and safety of individuals related to data. There is a possibility of government and legal regulations dictating the pace at which innovation and adoption takes place in this industry due to these ethical considerations.
References:
[1] R. Abduljabbar, H. Dia, S. Liyanage, S.A. Bagloee, Applications of Artificial
intelligence in transport: an overview, Sustainability 11 (189) (2019), https://doi.
org/10.3390/su11010189.
[2] M.N. Ahangar, Q.Z. Ahmed, F.A. Khan, M. Hafeez, A Survey of autonomous
vehicles: enabling communication technologies and challenges, Sensors 21 (2021)
706, https://doi.org/10.3390/s21030706.
[3] P.K. Agarwal*, J. Gurjar, A. Agarwal, R. Birla, Application of Artificial intelligence
for development of intelligent transport system in smart cities, Int. J. Transp. Eng.
Traffic Syst. 1 (2015) 20–30.
[5] H.E. B, N. H, Z. W, Security of cooperative intelligent transport systems: standards,
threats analysis and cryptographic countermeasures, Electronics (2015) 380–423.
[6] Briggs, B., Henry, N., & Main, A. (2019). AI fueled organizations. Deloitte
Consulting LLP. Retrieved October 31, 2019.
[7] B. Burmeister, A. Haddadi, G. Matlysis, Application of multi-agent systems in traffic
and transportation, IEE Proc. Softw. 144 (1) (1997) 51–60, https://doi.org/
10.1049/ip-sen:19971023.
[8] L. Chatterjee, C.m. Tsai, Transportation Logistics in Global Value Supply Chains,
Boston University. Center for Transporation Studies, 2002. Retrieved October 12,
2019.
[9] C. Cheung, Y.L. Chan, K.S. Kwok, W.B. Lee, W.M Wang, A knowledge-based service
automation system for service logistics, J. Manuf. Technol. Manag. 17 (6) (2006)
750–771, https://doi.org/10.1108/17410380610678783.
[10] M. Chowdhury, A.W. Sadek, Advantages and limitations of artificial intelligence.
Artificial Intelligence Applications to Critical Transportation Areas, 6-10,
Transportation Research Board, Washington, United States, 2012. Retrieved
October 24, 2019.
[11] M. Chowdury, A. Sadek, Y. Ma, N. Kanhere, P. Bhavsar, Applications of Artificial
intelligence paradigms to decision support in real-time traffic management,
Transp. Res. Record J. Transp. Res. Board 1968 (1) (2006), https://doi.org/
10.1177/0361198106196800111.
[12] R. Coppola, M. Morisio, Connected car: technologies, issues and future trends, ACM
Comput. Surv. 49 (3) (2016), https://doi.org/10.1145/2971482.
[13] J.A. Cortes, M.A. Serna, R.A. Gomez, Information systems applied to intelligent
transport improvement, Dyna 180 (2013) 77–86. Retrieved June 7, 2021.
[14] C.Y. David Yang, DL. Fisher, Safety impacts and benefits of connected and
automated vehicles: how real are they? J. Intell. Transp. Syst. 25 (2) (2021)
135–138, https://doi.org/10.1080/15472450.2021.1872143.
[15] C. Dirican, The impacts of robotics, Artificial intelligence on business and
economics, Proc. Soc. Behav. Sci. 195 (2015) 564–573. doi:10.1016/j.sbspro.20
15.06.134.
[16] emerj. (2019). https://emerj.com/ai-sector-overviews/ai-in-transportationcurrent-and-future-business-use-applications/. Retrieved October 20, 2019, from
https://emerj.com/.
[17] F. Lindow, C. Kaiser, A. Kashevnik, A. Stocker, AI-based driving data analysis for
behavior recognition in vehicle cabin, in: Proceedings of the 2020 27th Conference
of Open Innovations Association (FRUCT), 2020, pp. 116–125, https://doi.org/
10.23919/FRUCT49677.2020.9211020.
[18] D.J. Fagnanat, K. Kockelman, Preparing a nation for autonomous vehicles:
opportunities, barriers and policy recommendations, Transp. Res. Part A Policy
Pract. 77 (2015) 167–181. Retrieved November 28 2019.
[19] B. Fildes, M. Keall, N. Bos, A. Lie, Y. Page, C. Pastor, C. Tingvall, Effectiveness of
low speed autonomous emergency braking in real-world rear-end crashes, Accid.
Anal. Prev. 81 (2015) 24–29. Retrieved November 20, 2019.
[20] Gershgorn, D. (2018). Forget the space race, the AI race is just beginning. Quartz.
World Economic Forum. Retrieved October 19, 2019.
[21] H. Anandakumar, R. Arulmurugan, A. Roshini, Intelligent vehicle system problems
and future impacts for transport guidelines, in: Proceedings of the 2019
International Conference on Smart Systems and Inventive Technology (ICSSIT),
2019, pp. 1–5, https://doi.org/10.1109/ICSSIT46314.2019.8987831.