Istanbul’s traffic turnaround: using AI and analytics to decongest cities

By on 27/06/2023 | Updated on 27/06/2023
Istanbul traffic
Photo by Ikbal Alahmad via Pexels

Five lessons on using analytics, AI, and the Internet of Things (IoT) to understand and remedy traffic problems from one of the most congested cities in the world

When Istanbul leaders set their sights on deploying technology to improve transportation and transit, their city was ranked in the TomTom Traffic Index as the most congested city in the world. That was 2021, and the results of the growing problem have had a real, immediate impact on Istanbul’s residents, who can spend countless minutes or even hours to get from one neighbourhood to another – whether by car or bus. Istanbul’s traffic woes contributed to even broader, country-wide problems, including the high cost of importing oil to Turkey, an increased carbon footprint, and long commutes that decreased the quality of life for residents.

Traffic congestion is widely recognised as one of the most difficult city infrastructure challenges to address. Traffic problems can be capricious and unpredictable, changing significantly as the city itself evolves. Keeping up with current traffic challenges is difficult enough – predicting, anticipating, and mitigating future challenges is even tougher.

A grant from the U.S. Trade and Development Agency (USTDA), a government organisation that connects the US private sector to infrastructure projects in emerging markets, is helping change the equation for Istanbul’s traffic woes. The grant supported the installation of a SAS-based, AI-enabled analytics platform for evaluating Istanbul’s road traffic, optimising public transportation, managing traffic accidents, and adapting to changing traffic patterns. Using IoT-enabled sensor and networking technology combined with advanced analytics, this system is creating a real-time picture of Istanbul’s traffic conditions.

Which bus stops are contributing to the most significant congestion problems – and what are the likely costs of relocating them? How can public servants identify traffic congestion to dispatch teams to intervene in a timely manner? What are the optimal routes for reaching one neighbourhood from another – and how can current routes be shifted to enable them? These are only a few of the types of questions Istanbul’s planners and traffic experts are answering with the use of real-time data generated from remote sensors and analysed with SAS AI and IoT solutions.

Istanbul’s experiences have already generated a host of practical insights that can be used by urban planners, traffic experts, and other government leaders around the world. Here are some of the valuable insights uncovered in Istanbul to date:

Insight 1: The quantity of data collected is paramount in real-time transportation solutions

Data collection is the significant part of any analytic project, but the variety and volume of data is particularly important when forecasting congestion and transportation demand. The City of Istanbul modelled several different transport solutions, each requiring different types of data and with different data requirements.

For its traffic congestion forecasting use case, IoT sensor data indicates average vehicular speed for each road segment in one-minute intervals. By enriching this data with information about each road’s capacity, the relationship between the roads, and the number of vehicles on each road, the city was able to optimise its forecasting accuracy.

Insight 2: Run smaller and more easily achievable projects at the same time as tackling larger ones

Streaming data is valuable for forecasting transportation demands and optimising road and transit corridors. However, the data preparation and modelling activities can take a long time. While this work is being done, streaming data can be put to work to provide insights into citizens’ experience and behaviours.

In Istanbul, streaming data was used to start a Park & Ride campaign with the goal of increasing the use of the public transportation. The city implemented SAS event stream processing (ESP) and connected it with their popular mobile application which has over a million active users. With the use of ESP, Istanbul succeeded in identifying citizens that would be most interested in the campaign. The streaming data enabled the city to push notifications to them in real-time which resulted in an increase in people riding transit.

Insight 3: Many hands make light work… and many minds make more results

Understanding the potential for data has become essential for government employees. With tremendous value to be unlocked by data driven real-time insights, it is important for everyone to understand that collecting data and running streaming analytics is important. With a shared understanding of what analytics can accomplish, more use cases can be put on the table and the impact of a project can be magnified.

The Mobility, Smart City, business and IT teams in Istanbul conducted comprehensive training to increase the analytic maturity of its stakeholders. As a result, the city took on additional use cases and realised greater benefits.

Insight 4: Measurement of performance indicators plays vital role in justifying analytics work

One of the primary interests of an analytics project is the return on investment. For governments, the return could be saving money, reducing congestion, or protecting lives, to name a few. Planning to monitor various impacts as large, interconnected IoT projects can produce a variety of benefits.

The city’s Mobility team constantly monitors its impact based on a framework designated for the project to oversee greenhouse gas reductions, journey times, fuel reductions, total cost of ownership, running costs, and many other indicators. Monitoring these metrics enabled the team to determine that optimising bus routes decreased the number of hours in operation which resulted in reduced carbon emissions, bus driver needs, bus idling times, and costs.

Insight 5: Data can be leveraged by public servants for policymaking

The evidence that data provides is what public servants and elected officials need to set policy, make funding decisions, and enact laws. Incorporating data driven decision-making into the DNA of governments will improve how they use their limited resources to best serve their citizens.

Information from the mobility project is informing Istanbul leadership. Pedestrian accessibility to public transportation was analysed and the outcomes were provided to officials to determine the location of new bus stop locations and how to improve the effectiveness of existing infrastructure. Optimised public bus driver and bus assignments information is helping decision makers in day-to-day operations. Traffic forecasting enables the traffic management team to mobilise teams on the ground to direct commuters to low density routes in peak times. For example, the Mobility team used advanced analytics to predict with a 2.7% margin of error the rush hour traffic of one of the busiest routes in the city on the first post-pandemic day of school. This helped policymakers to take action on this particularly hectic and congested day to ease transportation in Istanbul.

Investments are already paying off

Istanbul is being rewarded for its concerted effort to use real-time data to decrease congestion and improve mobility. In the TomTom Traffic Index for 2022, Istanbul dropped from 1st place to 55th place. As more data is collected and analysed, the insights generated will continue to help Istanbul improve the lives of its citizens.

SAS Event Stream Processing (ESP) can enable you to analyse streaming data, uncover hidden insights with AI and make real-time, intelligent decisions. Find out more here.

About the author
Jennifer Robinson

Jennifer is SAS’ Global Government Strategic Advisor, working to help governments maximise the use of their data through data integration, data management, and analytics. Jennifer has a background in software development and local government. She co-wrote the book A Practical Guide to Analytics for Government and is featured in the book Smart Cities, Smart Future. In addition to writing articles and blogs about data-driven governing, she speaks to government leaders about emerging technologies and how to strategically adopt them.

About Partner Content

This content is brought to you by a Global Government Forum, Knowledge Partner.

One Comment

  1. Quamrul Ahmad says:

    I wish we can use the same for Dhaka city to manage trafficking more efficiently.

Leave a Reply

Your email address will not be published. Required fields are marked *