The results from cities using sensors, as reported at PIE 2014, were not quite the ringing endorsement that the manufacturers hoped for.
The most common sensor technology in use is the magnetometer, and the effects of surrounding infrastructure and fluctuations in local magnetic fields do not seem to have been fully understood, resulting in a loss of accuracy. Other technologies such as radar and IR are also available.
It’s suggested that where sensors are used to “zero out” single-space meters when a car leaves, some drivers have paid, only to see the meter return to zero due to a local fluctuation in the magnetic field.
Battery life is another issue. Early claims were for 7 to 10 years, or even longer. I understand that some sensors have stopped working after less than two years, and a safe average is closer to 3 to 4 years.
I am not quite sure how suppliers got this so wrong. Any city considering sensors should make the battery life a contractual requirement, with the whole cost of premature replacement born by the supplier.
The potential of real-time monitoring that detects when the space is occupied is clear. Occupancy data allow the city to fully understand use and to adjust parameters to optimize the operation. Availability data can be broadcast to help drivers to park without searching the area.
And by linking the sensor “read” to a payment, offenders can be picked up and ticketed with much greater efficiency than enforcement officers walking the streets.
I think we can get over-fixated on discrete data elements. Policies cannot be written and implemented on a single parking space; they are applied, at best, on a street-by-street basis.
Although clearly there have been issues about the accuracy of some of the sensor data, the SFpark pilot program has shown that these data are sufficient to refine charging regimes and, more important, oscillations closing toward an optimum solution. In some way, I think that this is the result from the San Francisco project — i.e., in a given situation, there is an optimum, and it can be achieved.
I am very, very nervous about broadcasting data to drivers and the suggestion that, in the future, drivers will use smartphones to identify and home in on vacant spaces. Here in the UK and in many other countries, it is an offense to use a handheld phone when driving, although for the present hands-free is still allowed.
Research here has shown that the use of a hands-free phone slows driver reaction times by an average of 27%, more than double that of alcohol consumption at the UK drunk-driving limit.
An increasing number of companies are prohibiting staff from using phones at all while driving, and it may not be long before any phone use while driving is banned, leading to the rather farcical situation where a driver will have to park to use an app to find a parking space!
For sure, if the sensor is accurate and connected with payment data, enforcement can become more targeted and more effective, but is this necessarily a wholly good thing?
As an industry, we are dumped on every time we write a ticket for a law breaker. If we start to detect and ticket 10 times as many criminals, there will be riots! We would have to rethink the whole process of ticketing, and be much more sensitive to the type of offense. A first-time offender, for example, or someone overstaying by a few minutes perhaps gets a slap on the wrist, whereas the serial offender or non-payer gets nuked.
Of course, if sensors were giving a 100% certainty of being ticketed for offending at a pay space and No Parking areas are still reliant on a passing enforcement agent, then serial offenders will quickly work out that the risk of a penalty is lower if you park illegally at the outset.
The law of unintended consequences can be a bummer.
So sensors can be useful, but I think there is still a ways to go, and as the people from New York might say, “We can do it all with a camera.”
Viva España
The first principle of parking management is to put in place controls that manage curb space and allocate the use between competing priorities to deliver the best outcome to the affected community. I am not sure anyone has written that down before, but that’s pretty much what we do.
Charging is secondary — to manage demand and to pay for the cost of operation. And ticketing is to deter non-compliance and, frankly, an admission of failure. Very few of us in the parking industry measure non-compliance in any reliable way, but the generally held view is that we hit about 1 in 10 offenders (although I remember research in London that showed less than 1% of offenses were detected, making illegal parking the cheapest option).
My colleagues in Spain seem to have finally got to grips with the compliance problem and come up with a spiffing new way of making parking work better.
Almeria in southeastern Spain has a population of a little under 200,000. Before 2012, the city was the definition of dysfunction. Surveys showed that only 6% of parkers were compliant, with 80% of drivers not even bothering to pay. Some 1,200 tickets were being issued each day, but there was no effective follow-up, the fines timed out and were canceled. Turnover in the limited-stay spaces was minimal, with fewer than two acts per day, with a 100% occupancy rate, and many vehicles parked for days at a time.
In March 2011, a new operator, the Empark Group subsidiary Dornier, which specializes in on-street related services in Spain, took over and has revolutionized the city’s 3,200-space system.
Occupancy is down to about 50%, with turnover at around five cars a day. Non-compliance is now running at 2.6%, meaning that there has been a 97% — that’s ninety-seven percent — reduction in non-compliance. (The bad guys are now using off-street garages, walking or traveling by bus.)
How has this been achieved?
First off, the system has been completely revised with a city Transport Plan, meaning that the controls are now up-to-date, relevant and coordinated with other parts of the transport system.
Second, enforcement is now better targeted. There are no sensors, but the parking meters require drivers to put in their license plate number, and Dornier patrols the streets using vehicles with LPR cameras. When the patrol vehicle sees a car not on the payment database, they contact a nearby foot patrol to ticket it.
Third, fines are resolved, rather than being put in a drawer and forgotten. About 12% are reviewed and canceled; everyone else is followed up, and virtually all fines are collected. All the systems used have been developed in-house by Dornier, and the philosophy and systems are now being offered across Iberia.
It seems to me that any municipality with a parking issue might just want to take a trip to Spain. PTT
The most common sensor technology in use is the magnetometer, and the effects of surrounding infrastructure and fluctuations in local magnetic fields do not seem to have been fully understood, resulting in a loss of accuracy. Other technologies such as radar and IR are also available.
It’s suggested that where sensors are used to “zero out” single-space meters when a car leaves, some drivers have paid, only to see the meter return to zero due to a local fluctuation in the magnetic field.
Battery life is another issue. Early claims were for 7 to 10 years, or even longer. I understand that some sensors have stopped working after less than two years, and a safe average is closer to 3 to 4 years.
I am not quite sure how suppliers got this so wrong. Any city considering sensors should make the battery life a contractual requirement, with the whole cost of premature replacement born by the supplier.
The potential of real-time monitoring that detects when the space is occupied is clear. Occupancy data allow the city to fully understand use and to adjust parameters to optimize the operation. Availability data can be broadcast to help drivers to park without searching the area.
And by linking the sensor “read” to a payment, offenders can be picked up and ticketed with much greater efficiency than enforcement officers walking the streets.
I think we can get over-fixated on discrete data elements. Policies cannot be written and implemented on a single parking space; they are applied, at best, on a street-by-street basis.
Although clearly there have been issues about the accuracy of some of the sensor data, the SFpark pilot program has shown that these data are sufficient to refine charging regimes and, more important, oscillations closing toward an optimum solution. In some way, I think that this is the result from the San Francisco project — i.e., in a given situation, there is an optimum, and it can be achieved.
I am very, very nervous about broadcasting data to drivers and the suggestion that, in the future, drivers will use smartphones to identify and home in on vacant spaces. Here in the UK and in many other countries, it is an offense to use a handheld phone when driving, although for the present hands-free is still allowed.
Research here has shown that the use of a hands-free phone slows driver reaction times by an average of 27%, more than double that of alcohol consumption at the UK drunk-driving limit.
An increasing number of companies are prohibiting staff from using phones at all while driving, and it may not be long before any phone use while driving is banned, leading to the rather farcical situation where a driver will have to park to use an app to find a parking space!
For sure, if the sensor is accurate and connected with payment data, enforcement can become more targeted and more effective, but is this necessarily a wholly good thing?
As an industry, we are dumped on every time we write a ticket for a law breaker. If we start to detect and ticket 10 times as many criminals, there will be riots! We would have to rethink the whole process of ticketing, and be much more sensitive to the type of offense. A first-time offender, for example, or someone overstaying by a few minutes perhaps gets a slap on the wrist, whereas the serial offender or non-payer gets nuked.
Of course, if sensors were giving a 100% certainty of being ticketed for offending at a pay space and No Parking areas are still reliant on a passing enforcement agent, then serial offenders will quickly work out that the risk of a penalty is lower if you park illegally at the outset.
The law of unintended consequences can be a bummer.
So sensors can be useful, but I think there is still a ways to go, and as the people from New York might say, “We can do it all with a camera.”
Viva España
The first principle of parking management is to put in place controls that manage curb space and allocate the use between competing priorities to deliver the best outcome to the affected community. I am not sure anyone has written that down before, but that’s pretty much what we do.
Charging is secondary — to manage demand and to pay for the cost of operation. And ticketing is to deter non-compliance and, frankly, an admission of failure. Very few of us in the parking industry measure non-compliance in any reliable way, but the generally held view is that we hit about 1 in 10 offenders (although I remember research in London that showed less than 1% of offenses were detected, making illegal parking the cheapest option).
My colleagues in Spain seem to have finally got to grips with the compliance problem and come up with a spiffing new way of making parking work better.
Almeria in southeastern Spain has a population of a little under 200,000. Before 2012, the city was the definition of dysfunction. Surveys showed that only 6% of parkers were compliant, with 80% of drivers not even bothering to pay. Some 1,200 tickets were being issued each day, but there was no effective follow-up, the fines timed out and were canceled. Turnover in the limited-stay spaces was minimal, with fewer than two acts per day, with a 100% occupancy rate, and many vehicles parked for days at a time.
In March 2011, a new operator, the Empark Group subsidiary Dornier, which specializes in on-street related services in Spain, took over and has revolutionized the city’s 3,200-space system.
Occupancy is down to about 50%, with turnover at around five cars a day. Non-compliance is now running at 2.6%, meaning that there has been a 97% — that’s ninety-seven percent — reduction in non-compliance. (The bad guys are now using off-street garages, walking or traveling by bus.)
How has this been achieved?
First off, the system has been completely revised with a city Transport Plan, meaning that the controls are now up-to-date, relevant and coordinated with other parts of the transport system.
Second, enforcement is now better targeted. There are no sensors, but the parking meters require drivers to put in their license plate number, and Dornier patrols the streets using vehicles with LPR cameras. When the patrol vehicle sees a car not on the payment database, they contact a nearby foot patrol to ticket it.
Third, fines are resolved, rather than being put in a drawer and forgotten. About 12% are reviewed and canceled; everyone else is followed up, and virtually all fines are collected. All the systems used have been developed in-house by Dornier, and the philosophy and systems are now being offered across Iberia.
It seems to me that any municipality with a parking issue might just want to take a trip to Spain. PTT
