Public transportation is a vital part of many people’s lives. Though we are currently experiencing reduced mobility due to the COVID pandemic, transportation agencies are working harder than ever to improve the infrastructure and provide timely, safe, and reliable service.
The agencies must face up to the challenges of reducing costs, increasing revenue, and trying to limit or act promptly when unexpected events, delays, or accidents arise. Moreover, they especially must consider the needs and expectations of travelers. Time is a precious resource for everyone: delays of buses or trains and traffic congestion in general often affects travelers’ plans. This is where Big Data analytics and predictive analytics come into play, bringing many benefits to the industry.
The predictive analytics applications in the transportation industry were widely discussed during the 100th annual TRB meeting in 2021. Insights gathered from Big Data could help transportation planners make more informed decisions when it comes to optimizing their services and improving the customer experience.
The 100th Annual TRB Meeting highlights
This year, traffic management became extremely challenging. Most of the fundamental theories on traffic management no longer apply. It is nearly impossible to develop long-term planning using these approaches as so many major traffic and transit trends have changed. We do not see the same peak hours, peak periods have shifted, weekday and weekend traffic became less distinctive, some facilities and areas have a different rebound, people’s attitude towards transit and ridesharing has changed. These are only a few examples of today’s situation.
According to Anita Vandervalk-Ostrander, one of the speakers at TRB, the industry requires new, more effective data sources and processing methodologies. The transportation experts should focus on the combination of data streams coming from
• Smartphone data
• Real-time passenger loads
• Traffic flow
• Transit ridership, Amtrak ridership, and ferry ridership
• Shared mobility data (Uber, Lyft, etc.)
• National transit database
• Real-time or near real-time counts
• Telework data analysis and reporting
By nature, both long and short-term planning efforts typically include a lot of assumptions. Nevertheless, in her presentation, Anita highlighted the need for reliable, continuous real-time or near real-time data streams and most importantly – education of transportation experts to work with the data:
“Data programs will suffer long term consequences as budget decision-makers don’t always understand how data drives decision making, but rather as a luxury or a federal requirement.”
Let’s dive into more details and explore some of the advantages of predictive analytics to the public transportation sector.
Solving promptly traffic congestion and issues
Predictive analytics can be applied to understand when and where congestion will be highest and react promptly in case of accidents or unforeseen events such as natural disasters or human events.
Transportation agencies can leverage real-time data to apply solutions that mitigate traffic congestion, for example, providing additional services during peak hours or offering alternative routes to avoid congestion areas. Travelers can be constantly updated on possible delays, route changes, etc. It can also be possible to redirect traffic or impose speed limits in bad weather, avoid or reduce vehicle accidents, or support management incidents when they occur.
Here is a couple of examples of how this technology has been applied:
· In Ireland, the city of Dublin, in partnership with IBM, has harnessed Big Data, using tools such as sensors, traffic detectors, GPS, and CCTV, to identify and solve the root causes of traffic congestion in its public bus network.
· In the US, the world’s largest community-based navigation and traffic app, Google Waze, has helped tackle the problem of double parking in Boston, Massachusetts. This is how it works: The city automatically ingests the data from Waze continuously building a database that stretches back to last year. The data is stored in a SQL database and analyzed using R, an open-source programming language. The data shows regular traffic flow and accidents reported by Waze drivers. For example, city officials would get notified about cars on the shoulder or double parking to be able act promptly.
Increase revenues and improve operating costs
Data can tell us lots of interesting facts to optimize costs and increase revenue, such as real-time traffic situation, the number of passengers taking public transportation or using certain routes, the peak days and peak times of traffic, etc.
It enables planners to better manage their services to reduce costs and guarantee a punctual and reliable service, especially during peak periods. With more advanced algorithms, planners can analyze different traffic scenarios during accidents, local events, or natural disasters. This explains how services may be affected and allow operational planning procedures to ensure traffic flow by controlling and offering alternative routes to avoid disrupting services.
Planning efficiently urban/developments projects
Finally, predictive analytics can benefit sustainability projects while ensuring smooth and safe mobility for all travelers. It allows for modeling the impact of urban development projects in the transportation industry.
There are plenty of application areas for predictive analytics for public transportation systems. It’s also not a problem to find data sources and reputable providers. The real challenge is to identify and start using efficient tools and methodologies to process, analyze, and exploit combined datasets for reliable forecasting. This way, governments, transportation agencies, and urban planners can make better, more focused future decisions.