For Whom the Bell Tolls

For Whom the Bell Tolls

What first comes to your mind when you hear the word “Bell”? A call to action, a clear signal that resonates over a long distance or even a warning? Examples that come to mind are church bells, fire alarm bells and the bells that go off when someone tries to violate a toll system. I would like to focus this article on the first definition of a bell as a call to action.

Smart Cities

I’ve been conducting some research into an emerging initiative known as smart cities and at first glance there did not seem to be a relationship between a smart city and toll roads. Then I started to think that toll roads were one of the first places where advanced technology was applied to transportation in the form of electronic toll collection. The advent of electronic toll collection has enabled us to drastically reduce the amount of land and infrastructure required to collect tolls and also provided some incredible flexibility for new payment strategies such as value pricing and variable tolling. As I started to address smart cities one of the first things that came to mind was a simple question – what exactly is a smart city? I find that thinking in simple terms is essential to the way my brain works so I came up with an interesting conclusion that a smart city is a city that when you poke it with a stick will respond accordingly!

So what exactly does this mean? It means that the smart city has the ability to sense changes in conditions within the city and in the surroundings that affect the city’s operation, develop strategies and then implement them successfully. Thinking this through a little bit more I realized that this could also apply to toll roads and that perhaps in addition to playing a significant role in a smart city, a really advanced toll road could act like a smart city in that if I poke it with a stick it will respond accordingly?

Smart Toll Roads

So let’s turn our focus to what exactly a smart toll road would be. Building on the definition of a smart city, a smart toll road would be able to sense changes on the toll road, changes in the surroundings that affect the operation of the toll road, develop appropriate strategies, implement them and monitor the effects. I think that because of electronic toll collection, toll road operators now have an advantage in the sensing department. Every one of our customer’s vehicles that is equipped with a transponder for electronic toll collection could also be considered as a probe from which data can be collected.

Subject to suitable constraints such as toll customers being able to opt in or opt out of the probe program and toll customer’s identities being protected, then this would be a really viable way to collect the data required for a smart toll road to sense changes in road operations. Transponder equipped probes would give us information regarding origin, destination and travel time. In this case origin and destination mean the entry and exit points to the tolling system that the driver used. Indeed, many toll collection agencies and toll road operators already make use of such probe data to provide speed maps to customers. But are we harnessing the power of this probe data to help us to define response strategies?

There may in fact be a limited number of strategies that can be implemented on the basis of changes in travel time to vehicle speeds, varying the toll is one strategy that comes to mind. So is there a way that we can get a more detailed picture of our customer’s behavior and what is currently happening not just in our network but on the surrounding network that has an impact and influence on what we do?

Data Lakes

Here again the work being done on Smart cities becomes interesting and very relevant to toll roads. A key part of the emerging smart cities initiative is the development of integrated data sets and the use of data analytics and discovery techniques to gain a deeper understanding of what exactly is going on. My way of thinking of this is that a data lake needs to be created that can act as a central repository for data from multiple sources. Certainly the transponder probe data would find its way into the data lake along with details of prevailing toll rates, data generated by other infrastructure-based sensors and in the near future data generated by connected vehicle probes. These would be private car, freight and transit vehicles.

There have been considerable advances in data science over the past couple of years, driven by Google and Amazon, to the point where a new way to approach data storage and manipulation is worthy of consideration. As you build your data lake and you find other data collection opportunities either from your network or from others then this new approach would mean that you would not question the value of new data. You would simply take it and put it into your data lake. In work done beyond transportation we have learned that seemingly useless pieces of data may well combine with other data in the data lake to create new insight into what exactly is going on. The larger the data lake, the likelier it is that this will happen.

However, creating a data lake out of currently available transportation data is only the beginning. The transportation data can be combined with credit card transactions, electricity consumption, retail activity and freight goods delivery for example, to create a multidimensional picture of the transportation network. Advanced data analytics and discovery techniques can then be applied to this picture to reveal trends, patterns and customer behavior.

So if we wanted to make our toll road even smarter than it is today what would be the first few steps to take? In order to support the approach I’ve outlined above then it is necessary to take careful steps towards the creation of a data lake. This involves acquiring access to a data platform, discovery platform and a data warehouse. You may have one or more of these already enabling you to embrace your sunk investment and make use of the legacy system. I think it would also be very worthwhile to conduct a quick review of the products and services available in the market. For example, these days there is much more choice with respect to either acquiring hardware and software. Of simply making use of it in the cloud.

Another important aspect in the preparations for getting value from the data lake lies in organizing for success. Effectively we need to build a bridge between data science and toll road operations. There may be people inside your toll road operation who are extremely skilled in information technology and have an in-depth understanding of your organization. These are likely candidates to harness the power of the data lake and deliver the insights.

It may also be necessary to supplement internal resources with some external specialists in data science and analytics. An awareness campaign among technical, marketing and operations staff regarding the possibilities of data science and the potential value that can be delivered by a data lake, might also be valuable.

This preparation to receive and to extract value from the data lake is absolutely crucial to success. As my granny used to say “you can be the best comedian in the world, but if the audience doesn’t understand you, you’re going to have a tough time”.

In many cases the steps towards the creation of a data lake will be the first few steps on a journey into the unknown. This gives rise to concerns regarding risk management and obtaining value for money. It may be very useful to consider a constrained, small-scale pilot to prove the value, build a business case for further investment and get your toll operations organization in good shape for the next steps.

I have come to realize that the application of advanced technology to transportation is also an exercise in change management. Successful application of the technology will result in change inside the organization. The change is usually a positive one enabling you to get even more value from the resources you already have, but may be initially viewed as a negative threat to the organization.

A really useful pilot would encompass some technical elements of technologies, products and services to be tested as you develop a business case and confirm that the advanced technology will deliver value that can be realized by your organization. It would also involve a close look at organizational arrangements required for success, by defining who would use the data analytics, who would use the insights regarding trends and patterns and how will this new information be incorporated into their jobs.

"As my Granny used to say...."

Another thing my granny used to say is “nothing happens until you make it somebody’s job”. So another essential element in a successful pilot would be to decide who’s job it’s going to be to operate the data lake, maintain it and extract value from it. Even to the point of developing job descriptions and job objectives that relate to the successful implementation of the data lake.

A bell works by propagating sound waves through the air and I expect that a data lake will also cause ripples of insight and revelation, each building on the other.

The inspiration for the title of this article came from a poem by John Donnne called “for whom the Bell tolls”. It’s useful to have the wider context of the poem as follows

"No man is an island, entire of itself; every man is a piece of the continent, a part of the main. If a clod be washed away by the sea, Europe is the less, as well as if a promontory were, as well as if a manor of thy friend's or of thine own were: any man's death diminishes me, because I am involved in mankind, and therefore never send to know for whom the bells tolls; it tolls for thee."

Toll roads are not islands either and we operate in a wider transportation context. So my advice is to turn your interest to the advances in data science that can be harnessed for toll roads and pay attention to the bell, it might just be a call to action.