T3 Webinar Question and Answer Transcript

Lessons Learned: Improving Reliability with Transit Signal Priority Systems -
King County Metro Transit and Los Angeles County MTA (January 22, 2008)

Back to Webinar Files


Q. Paul Olson asked: John needs to be clear that this is "point detection" It only provides information relative to the bus at a single point location at a single point in time and no more.

John Toone: That is correct. That's a good contrast between our initial system and the type of system where you can virtually set your point, and you have a stream of communication. That is a good contrast between our tag-based system and our current development.

Q. Joerg Rosenbohm asked: re: King County Metro: do I understand this correctly; you needed another controller in the cabinet that acts as the TSP generator? If so, how much did that device cost? Lastly, is the firmware proprietary to the vendor?

John Toone: Yes, and it isn't a second controller, but it is a second device in the signal controller cabinet. It is a transit priority question where in the cost of that is about 2200 to 25$00, and it is proprietary firm wear to the vendor, if you're outside the state of Washington, our initial contract made it available for free to anyone in the state of Washington.

Q. Wesley King asked: What is the length of express corridor John is speaking of?

John Toone: It is the Aurora corridor. It is about 10 miles, and there are 20 intersections.

Q. Domenic D'Andrea asked: What was the impact of your TSP to delay on the side street?

John Toone: That was one of the criteria that we had for our regional oversight committee was that to address that, and their input was that they wanted a little to no impact to the side street. They did not want to receive call from folks, so our -- and have that side street impact especially where you have crossing arterials. The goal was to have priority transit priority work so that it wasn't really obvious to someone who is sitting at the intersection, so we achieved that end goal. We've had very little impact and very few calls about signal priority impacting general purpose side street traffic.

Steve Gotas/Reinland Jones: We proceeded in a very similar fashion, and again it was the traffic engineer's side over here that was primarily concerned about the impacts on the side street. Working with city of Los Angeles and the city of Glendale, and utilizing similar signal priority rules that they had implemented, it was determined by our advisory committee that we didn't need to actually measure it because those two projects when they conducted their evaluation really didn't come up with any noticeable impacts on the side street, and our jurisdictional partners were all in green light that we would handle cross-street traffic impacts on a case by case basis, so if I had to just throw out a quick and easy answer, it was that we didn't have any noticeable impacts.

Q. Ben Baldwin asked: Please clarify? How many signals in this corridor?

John Toone:  That was 20.

Q. David Tomzik asked: Please detail the conditional priority, if schedule adherence was one of the factors, detail how the threshold was decided. Also, was threshold revisited after evaluation?

John Toone:  We included part of our capabilities of our system is to do on-time performance as one of the conditions, but actually we haven't used it as much partly because the buss themselves don't know whether they're on time so we have to calculate that on the roadside when we're setting up our eligibility tables, but also because we found from our evaluation debt having predictable trip had a better impact than trying to speed up, so we've actually gone through and our TSP policy is when we select a route, a specific service route that we try to give priority for every trip in the entire trip to give it a predictable travel time down the corridor, but we are moving now forward that we may given that we have more control over the logic, our evolved working relationship with the traffic community is that we have what we call a mother may I system, how much we can get, how much interruption to the general purpose signal timings we can do, and we decide how we spend that out. We use most of our conditioning to make sure that we are getting as close to the maximum amount of priority as we can to benefit transit.

Q. Ben McCawley asked: Is the controllers TS1 TS2 or 2070?

John Toone: Well, one of our criteria was that we needed to work with many different controller platforms, so we have -- TS2 and 2070 we were -- we have the CPAC firm weary con owe light, ASE 2 and ASE3 firm wear as well as 2070 I think is next phase.

Q. Cade Braud asked: Any maintenance issues related to the RFID readers?

John Toone: Yes. When we went into our initial system, it was important to us to use technology that was similar to the traffic controller industry so we were using technologies that were comparable to early 90s traffic controller equipment, so we -- we had to replace the EPUs in our readers because the shuttle program bought every 8088 processor in the United States so we needed -- that was a significant issue is we had to change some of the platform on that. Also the RFID requires a directional antenna which they get bumped around and require some aiming. There is RF interference. We will to change the frequency on a good number of our equipment because it was interfering with the meter reading equipment for the gas and electric readers in the King County area in certain locations, certain frequencies. And then of course every one of those is an individual device, and having lots of individual devices out is more points of failure, so it adds to the complexity of the system as a whole.

Steve Gotas/Reinland Jones: Sure. From a maintenance perspective on our side, you know, of course with any network system you're going to have some equipment systems failure and some network instability, but what we've seen is we moved into kind of a second generation of our network configuration. The majority of our maintenance is relatively minor. We had to do a little power conditioning and power cycling when our hardware failures, but the only time that we have any major problem sincerely when we have cabinet and pull knockdowns. Then on the bus side we do have a central monitoring system that allows us to take a look at the on-bus systems are performing, and we do have a maintenance contract with our vendor who is responsible for going out there and resolving any issues. Again, those aren't above and beyond any other typical on-bus systems maintenance requirements, so we're pretty comfortable with those costs at the moment.

Q. Ben McCawley asked: What type of central system software are you using?

John Toone:  I am assuming the question is about signal priority central system software, and we're using one that was developed by the vendor, and we also have developed several in-house we have developed several system to say help us do O&M tracking, performance evaluations, and such like that in case they mean what type of controller, we've worked with icons and not coming up with the other one at the moment so I will turn it over to L.A.

Steve Gotas/Reinland Jones: Sure. We don't have a typical central system application outside of what we use for operations and maintenance which more than anything else those are simple network monitoring tools, and we do have interfaces with, lease see, quick net 4, you can also see some of our information on city of L.A.'s ATCS central control package, but for the most part our system from an operational standpoint everything re sides on the bus or at the intersection.

Q. Kevin Hanson asked: Has John considered other technologies that are present in the area?

John Toone: Yeah. You might be alluding to early on Pierce transit was part of our regional oversight committee, and they early on decided to go with an opt con based technology, and we considered that at the same time, but it wasn't going to work for the needs of the traffic controller traffic jurisdictions in jurisdictions in King County, and it wasn't going to work with the profile for how we operate our transit where we have -- we don't have dedicated vehicles to route and we have multiple routes of the same corridor. We needed to have something that could transmit a message to the roadside and was flexible that a different vehicle could do the same route.

Q. Mark Cassel asked: What was the source of the funding that Metro distributed? Did any of it come through CMAQ?

John Toone: Our initial system did have grant funding, and I believe it was at least -- it was in part CMAC. I think we had other funds, but it was like I said early 90s, so 15 years ago.

Steve Gotas/Reinland Jones: We distributed primarily local money. I am assuming maybe this question is talking about our other jurisdictional implementations we funded, and that was all local money. We've only received to date a very small portion of CMAC money for our own internal signal priority project, and I would say of the 20 million that we have allocated about 25% has come from CMAC source.

Q. Paul Olson Asked: How do you know if the bus actually receives and is able to make use of the priority service interval? Does the bus pass thru the intersection during any granted priority interval? How do you really know?

John Toone: Paul always has the tough questions here. It is a good one in that that was the work that we were doing with I cons [ indiscernible ] is to get the feedback of the logs from the signal controller system to say log what they did with the signal priority call that they got it and to find out how they acted on it, but our own equipment does log in what requests we made and we monitor the phase states at the intersection well, as we're moving through this, so we're able to with our own back office systems go through look to see what was taking place and what happened in time nearby that so we're able to go through and see what kind of benefit we believe that we got now that's working with traffic central system sincerely going to give us a better assurance that something is happening on that.

Steve Gotas/Reinland Jones: Sure. We probably do an inadequate job measuring how affected we are, taking advantage of priority out on the corridor. What we try to do is keep everything at the field level to the degree possible. We do have logging capability where the traffic signal controllers send a data packet back to a central logging over here at metro. We haven't had the opportunity to date to take a look at that information, but the way we try it get around this is -- we actually send an update message as the bus gets close to the intersection and the goal here is to let the intersection know really whether or not we're arriving according to our original estimated time of arrival, and then we check out of the intersection as we pass the mid-pointed of the intersection so the intersection can free up any additional time that it had granted to our bus. How do we really know? We're going to have to go and look at the log information and compare it with our on-bus data. We really don't have any time line on when we would start doing that type of analysis.

Q. Paul Olson Asked: If the bus is granted priority and actually passes thru the intersection during the "normal green interval do we claim that TSP has helped? What about the additional delay because the priority interval has essentially passed by as unused?

John Toone: When we did our software updates that's one of the questions we wanted to tackle on that, and the programmable logic and the priority generator has allowed us to monitor the states, monitor if the priority phase is green, and then actually not request priority, and in fact that logic has gotten sophisticated enough that we'll also know that the signal is about to go green, and that way it is not a good use of priority, so that's how we maintain getting the most out of the amount of priority that is granted is that we put the intelligence on whether or not to even make the request, so thinking back to Paul's question on this is that we don't ask for priority on the normal yen and then we log -- green and then we log that we got a call and did not ask for it because it was already green.

Steve Gotas/Reinland Jones: Let's see. The way that we approach it is a little differently. Since we make the decision to request priority on the bus and the bus has no idea what phase the intersection is, we go ahead and make that priority request to the intersection and allow the traffic firm wear to take over and decide how it wants to implement priority. If the intersection decides that when the bus is going to arrive there is already going to be a green phase, then it will go ahead and disregard that priority request and continue to operate as it normally would as if it didn't have a request in the queue, so then the question as to whether or not we would assume that TSP helped if we approach during a normal green obviously we would look at it from the standpoint that that wasn't a benefit from TSP.

Q. Sandra Majkic asked: Is there a team or task force formed to include traffic engineers and planners to work closely on TSP implementation?

Steve Gotas/Reinland Jones:  Sure. At the on set of our project back in late 1998 we convened a 22-member advisory committee to guide TSP implementation, and this team was comprised of a number of local traffic engineers from some of the jurisdictions within the L.A. County, other transit operators, also included a wide mix of management and technical staff. That team has since been dissolved, but we do have a fairly active body out here in L.A. County of traffic engineers who attend regular meetings, and what's more of an informal process than anything else.

John Toone: Yeah. I think the question was actually are there planners, and there are planners of TSP, but no originally we did have a formal process, our regional oversight committee where we had traffic jurisdictions representative from all the local traffic jurisdictions, and we went through I phase where it was much more informal because we had our operations and maintenance agreements and did a lot of one on one work with the jurisdiction that is we had equipment installed in, and then now we're swinging back around to the local MPO is forming a regional signal operations committee, and signal priority now is a mature enough along that it is one of the top ongoing topics that they address.

Q. Sandra Majkic asked: What TS display is being used?

John Toone: I am really not sure where the question is going.

I am not sure what TS display is?

Steve Mortensen: Transit signal? Photographic signal -- traffic signal? Okay. We'll go to the next question.

Q. Sean Kennedy asked: How did you prioritize corridor implementation?

Steve Gotas/Reinland Jones: With our original pilot demonstration we went through a fairly exhaustive corridor selection process, and I think we looked at our 25 high eases corridors, and then we selected a number of attribute to say use to evaluate the corridors, things like what our operating head ways were, how behind schedule buss were, the corridor length and also the number of jurisdictions we went through. The process for determining what corridors we roll out now is has become much easier for us as a group. We have a metro rapid five-year implementation plan that calls for the deployment of 26 pseudo BRT lines within the County. We now implement according to that metro rapid implementation plan, and again those are primarily focused on putting pseudo BRT services on the high rider ship corridors.

John Toone: Initially -- initially we had fairly formal process of evaluating corridors, and we looked at how we came up with reign corridors initially and after those we moved to where signal priority was just one tool for transit priority in general we used, and on all of our projects we looked to see if signal priority is appropriate, and we have tools such as a planning model that does a cost benefit and let's you play with some traffic timings and costs and look at the scheduled service to see if signal priority is going to have a positive cost benefit.

Q. Peter Koonce asked: Has anyone else other than LA MTA implemented a similar Wireless TSP System?

Steve Gotas/Reinland Jones: We here at L.A. metro have actually granted funds to two other trance it'd operators in the region, foot hill transit and Long Beach transit who are going to be implementing basically an identical system to our wireless setup. Outside of that, I don't know anybody else who's done it within the country although we have been getting a significant number of information requests in the last two years, so I am assuming there are other folks looking into it.

John Toone: We went to L.A. because we knew they had one installed. I hear that Cape Cod has a few buses running on one, but I don't think I would characterize it the same way. Everything is looking to L.A.

Q. Yadollah Montazery asked: Does signal priority work best with fixed time or actuated signals?

Steve Gotas/Reinland Jones: For L.A. we actually go through both fixed time and actuated signals, and I am not sure we have ever measured a difference in how well priority works at the different types of locations. I would say we've had less problems implementing at fixed time locations, and we have -- but we haven't run -- along the same note we haven't run into any significant problems when we've dealt with ac you waited signals.

John Toone: We actually don't work with fixed time signals very much. Most of the fixed time signals in King County are in the downtown Seattle area, and for the most part we don't do signal priority on those, especially they're typically here used more on very high-volume corridors where you're not -- there is hard-to get-time out of it, so we do target towards ac waited signals primarily.

Q.  Peter Koonce asked: Have you had any problem with signal interference to the wireless system?

Steve Gotas/Reinland Jones: We generally design the systems with the understanding that we will have some signal interference, so the signal is basically designed to accommodate interference. In terms of post installation interference, we really haven't had any significant problems. I don't think there has been a single incidence when we've lost connection with the bus because of another signal impacting ours.

John Toone: Well, we don't have the wireless system, but we have had some rounds with just RF interference, and in fact we go through a fairly rigorous RF survey just to see what the environment is before we go in and do a signal priority.

Q. Ben McCawley asked: What access point system are you using?

Steve Gotas/Reinland Jones: We use a Cisco internet 1300 bridge right now as our access points, fairly robust external 80211 access pointed antenna and access point and single enclosure you're, and it is all done through power over Ethernet.

Q. Kevin Hanson asked: How many intersections use this technology now?

Steve Gotas/Reinland Jones:  We have about I want to say 95 intersections up and running at the moment. We have another 20 intersections that will go online shortly, and our final build is going to be upwards of 400 intersections.

Q. Peter Asfour asked: Any issues with security of the wireless system?

Steve Gotas/Reinland Jones: I would say no. We haven't had a single intrusion to date. That's not to say that wireless security isn't a major issue. We're currently in the process of enhancing our wireless security, but basically we're using pretty standard 80211 security protocols in our instance we're upgrading to a higher level of WPA.

Q. Eddie Curtis asked: Does the decision to grant priority consider the demand on phases that will be truncated to provide the additional green time to accommodate the phase serving the bus? Have any studies been done to assess these impacts?

Steve Gotas/Reinland Jones: With some of our more advanced system integrations, for example, when we interface with the city of L.A., those types of issues are considered and signal priority won't be granted if there is a very heavy demand on a particular phase, but at most of our with most of our jurisdictional partners since we don't have centralized system monitoring things and also we don't have sufficient amount of logic in the traffic controller to date built into the system, those types of phase demands are discounted at the moment, and again we only play with 10% of the signal cycle, so we haven't seen that much of an impact on a cross-street.

John Toone: This is similar to the side street question that we had earlier. We addressed that in the design of the signal controller timings, and we don't have designed in signal priority plans that is going to have a significant amount of impact on the side streets, so all of our plans made sure that there is -- we never used phase skipping, and there is always a minimum amount of green that's required on that.

Q. Sandra Majkic asked: What controller performs the best?

Steve Gotas/Reinland Jones: Sure. I don't think any one particular controller performs the best. We've had success with the Type 1 70s and newer type 170 runs HTH C 11 engine boards, also the 20 p 0s and the ASE controllers. I think from a visual perspective since a lot of what we do is out in the field, we have a much easier time seeing how things are performing utilizing the viewing capabilities at the 20 2070s and the ASE controller provide but from an actual -- the actual priority function itself really seems to work the same across all of the different controller platforms.

John Toone: We don't have a single controller that we say use this one. There are several that we prove, but we do have criteria for controllers. I think I said before we worked with light and eagle and 2070. Each of those has a controller that does signal priority functions in it. It is beyond meeting our criteria for working with the signal priority system. It is very much up to the local traffic jurisdiction of what controller are they familiar with, what controller gives them the priority handling that goes to their philosophy, but there is several that work equally well.

Q. Kevin Hanson asked: Can this system also work on high priority? Has it been installed in that fashion?

Steve Gotas/Reinland Jones: The system wasn't design to do operate on high priority. It does have the capability, but it is something that we would probably never want to experiment with here in Los Angeles County simply because of the back lash that we would anticipate. So I don't think we have any long-term plans either to use the system to do anything like vehicle pre else or specifically emergency vehicle pre else. Right now we operate completely in a low priority mode.

John Toone: The technically we could wire in our requests to the high priority inputs, but for the same reasons we don't. The city of cattle is working with -- city of cattle is working -- put in a transit priority routine that is a blending of the two and can handle light rail and CRT and transit signal type priority requests and sort of a scaled approach where it will do high priority if it can or something similar to it but then scale back to the more conditional on low priority.

Q. Wesley King asked: Do any of these corridors function in HOV or dedicated bus lanes along with TSM?

John Toone: Yeah. We have transit lanes in some of our corridors, and we do use signal priority along with bus lines. There is quite a bit of discussion about that around whether or not you really get benefit out of it. We believe that you do get benefit because of the reliability aspect that signal priority gives you.

Steve Gotas/Reinland Jones: We haven't to date used our system on a dedicated bus lane or for HOV. We do have an implementation here in Los Angeles. Our orange line operates on a dedicated bus line but that utilizing the city of L.A.'s loop system. We have a number of studies moving forward now assessing whether or not we should convert our mixed flow lanes into dedicated bus lanes, but we haven't seem to get the public traction we needed to get these converted.

Q. Yadollah Montazery asked: What is the average cost per signalized intersection including planning, design and implementation?

Steve Gotas/Reinland Jones: Sure. I will throw out very general numbers because it will really depend upon the number of intersections and the number of buss. We realize significant cost savings by expanding the scope of our projects to incorporate more buss in more intersections, but in general we're talking between 10 to $25,000 per intersections, and that that depends on the specific configuration of the intersections as well as the type of controller in place and the type of cabinet and a host of other physical characteristics, and then it is also about 15 to $20,000 per bus. We're hoping to say reduce our on-bus costs by integrating some of our on-bus systems into a single entity as opposed to utilizing our own black box solution but we haven't gotten that far yet. I would throw out there anybody is going through a process -- a cost estimating process for system similar to ours, I would go ahead and give me a call and I can help you work through that a little bit.

John Toone: Our existing tag based system is a little more than that. It is the raw equipment itself is about 30 to $35,000. That's sorted of assuming normal amount of installation and costs and whatnot, but along with the same caveats as no new signal controller cabinet and whatnot, major infrastructure is needed, although after planning everything, it is probably closer to 50,000 per intersection or so. With our system we're developing now, the on-board part of the system is going to use the same equipment as an existing fleet wide so marginally for the TSP system there will be none on board the bus, and we're hoping to eliminate most of the roadside costs to where we'll have just $2500 or so piece of equipment in the signal cabinet plus the communications trunk behind it, so probably fairly similar to the L.A. cost.

Steve Mortensen: Thanks. I would like to add that the FTA and ITS Joint Program Office sponsored guidance documents, which I spoke of at the beginning of the T-3 webinar, that also provide cost information, so I wanted to let you know that information is available.

Q. Sadiq Pirani asked: What other measures have implemented? Dedicated bus lanes, queue jumps, etc.?

John Toone:  That goes along with we have a whole package of transit priority treatments besides signal priority and we have done HOV lanes. We haven't -- we haven't used a dedicated transit-only lane in conjunction with TSP. We do use Q-jumps and we also do some quite a bit of spacing stop location work as well.

Steve Gotas/Reinland Jones: We're in a similar situation. Although like I said earlier we haven't implemented any dedicated bus lanes the corridors we implemented priority on. We don't have any Q jumps. We have increased our station spacing to about a quarter of a mile to about three quarters of a mile between stops, and we've also moved all of our stops far side but other than that we haven't done anything too fancy.

I actually like that. Add onto that that we've actually in getting into signal priority and getting into the controller realm is that we've funded quite a bit of just signal timing regardless of whether or not it is signal priority, but well-time, having a well-timed corridor is we've actually let out grants at local jurisdiction to say just do signal timing period.

Q. Eddie Curtis asked: What are the day to day maintenance and operations staffing requirements that keep the system functioning?

Steve Gotas/Reinland Jones: We have right now a maintenance contract with our system supplier or our systems integrator. There is a day-to-day monitoring of the network, and that's done remotely utilizing a series of cellular links, and then we also monitor our on-bus systems through a wireless connection at the division yard. We don't have a very robust maintenance requirement. It is specifically when something fails the contractor needs to go out and remedy the situation within a reasonable amount of time, generally within one week, but we haven't been able to get a good handle on what the maintenance impact is going to be on our agency when we bring the may maintenance in-house, and there hasn't been that determination made either, so those are some of the questions we're going to be struggling with probably in the next one to two years.

John Toone: We have one to 1.5 FTEs that do system monitoring. We also have a small contract with our vendor to do some trouble shooting calls, but mostly our field maintenance is done through interposal agreements with local traffic jurisdictions, and when we monitor centrally that there is an issue, then we issue trouble calls out to the local traffic jurisdiction, but like L.A., we are looking at what's the impact if we can bring all of that and centralize the maintenance in-house.

Q. Kevin Hanson asked: Is this technology patented?

Steve Gotas/Reinland Jones: No, we haven't filed any patents on any of this, and the way that we view this is it is not a product but more of an approach, and so we're not in the business of patenting anything, and we do own our system and all of the work that our system suppliers provided to us, and so they don't hold patent rights either.

John Toone: The equipment we currently use is provided by a vendor, but it is not a pat noted approach. There could definitely be someone else developed a similar system on that. Then I believe at one point there was a patent held by 3M for using differential GPS, but I believe that that patent expired and they never really actively pursued defending it.

Q. Mark Cassel asked: Were any of the jurisdictions called upon to provide any match dollars for the implementation? If so, was there resistance?

Steve Gotas/Reinland Jones: No. We fully funded this project. There were no hard match requirements from any of our jurisdictional partners. We did ask that they provide some staffing resources for plan reviews and also some assistance when we're out in the field during doing installations, but we didn't meet up with very much resistance at all. Any resistance that we did get from the jurisdictions has been more based more on technology concerns and legal issues than money.

John Toone: Typically we're going -- we usually fund our own -- the signal priority installations and a lot of time they're done in concert with a larger project, and in many cases we're partnering into a local jurisdiction's grant application to either the state or federal grant, and then our TSP becomes some of the local match.

Q. Cade Braud asked: What has happened to the TSP equipment on the corridors that were initially implemented in the Metro Rapid program (e.g. Wilshire, Whitter)?

Steve Gotas/Reinland Jones: That system actually still is in existence and still in place on both WHITTIER and Wilshire, and it is actually expanded to another 15 or so corridors over the last three years, so in L.A. there has been some confusion between the two systems. Typically within the city of Los Angeles which again like I alluded to in our presentation makes up about two-thirds of our service area, we deploy city of L.A. loop com km, and in areas outside of the city of L.A. or for quarters transverse through we deploy the wireless system so there is two systems in use throughout the County right now.

Q. Mark Cassel asked: Were there any situations where a non-linear closed loop signal system complicated implementation?

Steve Gotas/Reinland Jones: SI will answer this short. We haven't come into any complications associated with the issues like that.

John Toone: None specific. We did have complications working with one central system, one city had a central system that tried to do some minimal amount of adaptive type, and we were not able to get priority working with that system, but that was relatively minor, just one instance.

Q. Ben McCawley asked: What steps has your taken to ensure that your equipment is not susceptible to ARP storms caused by i-phones or other wireless devices?

Steve Gotas/Reinland Jones: We haven't hardened any of our equipment or our network above and beyond the basic -- above and beyond what the hardware that's available for the program, but again we haven't really seen any instances where somebody has either tried to hack into our system or where we've been receiving interference from other devices, so we're going to continue to monitor things. If we start to see instances of intrusion, we're going to of course take the appropriate steps, but until such a time, we're really relying on the functionality of our hardware to protect us from a lot of that.

John Toone: We're still in the design, so I am listening eagerly to L.A.'s answer on that.

Q. Domenic D'Andrea asked: In LA, what agency permits the signals? What agencies own and maintain the signals?

Steve Gotas/Reinland Jones: In Los Angeles the owning agency is generally the jurisdiction within which the signal resides. I think that's in all cases. In terms of owning and maintaining, there is some separation in that our L.A. County department of public works does provide a significant amount of maintenance support to our local agencies. Of course with 89 jurisdictions we have other agency that is maintain their own signals as well, and then we have instance where big entities like the city of Los Angeles has maintenance contracts with some of our local agencies. There is a good mix of who owns, and I am sorry there is a good mix of who maintains. In terms of ownership it really is based upon where the intersection lies and we do have quite a few instances of joint ownership the County where up to four different jurisdictions might own a single intersection.

Q. Mark Cassel asked: Were there any champions from outside the transit agencies that helped drive the implementation process? Who were they and what roles did they serve?

John Toone: I would say we are primary champion for signal priority. Many times not necessarily from the traffic engineering group within the local cities but many times they city planning or city counsels will have their own policy goals that are directed at improving transit and those tend to be help drive from the local jurisdiction side getting signal priority implemented.

Steve Gotas/Reinland Jones: We're in a similar boat. We didn't have any one particular champion for our project. At the on set of our project we had a lot of support from a variety of city managers, CEOs from other transit properties and of course very influential traffic engineers within the region, but more than anything it was really a collaborative effort, and there wasn't any one individual driving the process.

Q. Domenic D'Andrea asked: In King County, what is the name of the system software that was used for TSP?

John Toone: I am assuming that question is about the TSP system software, and the vendor is McCain traffic systems, and their TSP system was developed in conjunction with us.

Q. Justin Begley asked: Was traffic modeling needed to demonstrate minimal impacts to cross streets? Where can we find statistics that demonstrate improvements to cross street traffic LOS?

John Toone: We used quite a bit of traffic modeling. We have several people within our work group here in transit that do traffic modeling. We typically we always do sync row modeling for every installation onto help determine what the signal time will be and what they need to be to justify the cost. We also for corridor and more complex we use VISM modeling as well. For planning level we use an in-house developed tool, the TSP interactive model which is do the cost benefit planning on that to decide to look at the intersection between number of trips and timings and whatnot.

Steve Gotas/Reinland Jones: Let's see. Here in Los Angeles we haven't had to do much modeling at all, and we definitely weren't able to measure cross-street traffic impacts. Since our system is a requirement of our metro rapid program, we really viewed it as an essential element as opposed to an optional piece, so that's really alleviated some of the burden of doing extensive modeling marshal if there is any way. A good place to find that type of data, however, might be from the city of Los Angeles who did conduct some statistical analysis of cross-street traffic impacts on some of their corridors fortunately I think they selected -- they did an analysis primarily of the WHITTIER corridor, so anyone interested in that can also go ahead and get in touch with me and I can get you to the right person at the is city of L.A.

Q. Sandra Majkic asked: What is the reduction in cost by removing readers?

John Toone: John Toone: The individual reader, the equipment cost is 8 to 9,000 for a single reader alone plus installation on it, so we're expecting that it is going to reduce our installation cost like 70%.

Q. Domenic D'Andrea asked: For either LA or King County, were there any issues with acceptance by union drivers? Any union issues?

Steve Gotas/Reinland Jones:  We've had very few problems with our unions in terms of both our operators as well as our mechanics. We did some initial outreach with the unions, and there was seemed to be a general comfort with our system that we ran into a couple of instances where we had bus operators disabling systems simply because they were unaware of what we were doing with our GPS data, but as we moved forward we haven't had any issues with our unions, and I will knock on wood right now because I know there is always things that can creep up, but even from an installation standpoint our unions have been pretty comfortable with us bringing in outside crews to perform that work.

John Toone: We've had no significant union issues. Some union related considerations we had were that we did not wanted to have the on-board equipment did not -- we did not want to have additional displays or additional interactions for the driver. We wanted to keep the operating environment in the coach simple. We didn't want to add to the confusion on the bus. From the roadside maintenance of course there is the maintenance issues of who gets the piece of work to go into the cabinet and do the piece of work, and there really wasn't an issue about it, but it was just a consideration of acknowledging that this was -- that once we selected it was going to be a precedent that would carry forward probably to other jurisdictions.

Q. Yadollah Montazery asked: Any impact on pedestrian wait time to cross the intersection?

John Toone: The pedestrian cross-times are another thing that we looked to protect when we were doing the signal priority timing so be sure that we don't impact the minimum crossing time. I will acknowledge in some cases we do end up shortening them down slightly. For ones where we don't reduce it past the minimum, but if there is an exceptionally long one, we can shorten what is the apparent crossing time to some people who use crosswalks frequently.

Steve Gotas/Reinland Jones: We do it exactly the same way. I would just did just DITTO what John just said.

Q. Mark Cassel asked: Are both systems based on conditional priority? Was there any shortening of run times following implementation that would in turn drive additional use of the conditional priority system?

Steve Gotas/Reinland Jones: You know, we at the moment have reconfigured our system so that we request priority all the time. We have changed operations from schedule base to do head way based. In terms of shortening runtimes following the implementation, gem what our scheduler do because we do have time points for the beginning and end of our lines, is that they'll accommodate for upwards of up to a 10% reduction in their typical round-trip running times, and once we modified our station spacing they might increase that to 15 to 20%, so we have made some changes, but again we request priority all the time at the moment, and we really allow the intersection to take over control and to determine how they want to implement the priority. Some of them do implemented conditional priority and others grant us priority all the time, so it really depends on the jurisdictions that we're traversing through.

John Toone: I guess I would say we do call what we do conditional priority because we place conditions on all the calls although I think this may mean on time using on-time performances that condition of whether late or on-time, and we don't use that necessarily, but in terms of reducing the runtimes, we do actively try to work back with the scheduling group to make sure when we do an implementation on the door that we press to take time back out of the corridor because I think most operators know it is hard to get the feedback that you could operate faster on a schedule-based service, so we need to actively try to press the schedules to take the time out.

Q. Sandra Majkic asked: What is the highest pedestrian volume at the corridor crossings?

John Toone: I don't think I have that off the top of my head. We have a wide variety where we have some corridor that is have very low pedestrian crossings that are state routes and whatnot up to we have some that are in corridors that have a fair amount of pedestrian traffic.

Steve Gotas/Reinland Jones: Probably want to answer the same way. I don't have the actual numbers, but we do go through some very heavily pedestrian oriented intersections where we're getting calls continuously. I can probably get that information fairly easily, so if -- Sandra, if you're interested in that information, I can try to do a little digging for you to get a little more concrete.

Q. Ben McCawley asked: On the WiFi If the authentication method goes down, what is the impact on the system? What are the consequences of authentication failure?

John Toone: Well, if our authentication goes down, then essentially we're not going to be able to get messages from the bus to the intersections, so I would say that it would be a catastrophic system failure until we got it back up and running. There will never be a time when we supersede any of the security requirements that is we suspect in our system design, so I guess that's probably the best way to answer it.

Q. Joerg Rosenbohm asked: Re: Seattle: are you writing your own firmware to accommodate these other functions that the vehicle can send to the Central system? Are there standardized interfaces to the various vehicle subsystems?

John Toone: Yes. We actually at the same time we were going through and developing the TCIP standards group was active and we contributed our list of 35 different variables related to signal priority to the TCIP standards group, so there is a nice feedback for us on that in terms of piece of information we can look at. Then we aren't necessarily specifically writing our own firm wear into our signal into the TPRG, but we do write our own logic into that, and we have a standard that we use but we're also can adapt it to use any number of information, but we are looking at using that to do head way management on time management, and use any of that information that can come from the bus. When we move to the message-based system using the Wi-Fi, we're going to transmit as much information as known to the bus to the roadside so that it is available to use for determining priority.

Q. Kamil Kaluski asked: Are you concerned about WiFi security issues?

Steve Gotas/Reinland Jones: Yeah. We've been concerned with Wi-Fi security issues since before we even started the process. I guess the key really is when we discuss wireless security issues is what are we protecting? What is a reasonable amount of security? We're currently in the process of like I said upgrading to WPA level of 80211 security, and then the remainder of our system is fairly secure and by that I mean our systems, our wireless systems at the operating division are already much more secure than our on-street networks, so we know it is a concern. It is definitely something that comes up at all of our jurisdictional coordination meetings, and I think we just need to stay at the cutting edge of wireless security to make sure everybody is comfortable.

Steve Mortensen: Thank you very much. I have a disclaimer here that I would like to say. While the presenters answered questions regarding specific technologies and equipment and their usefulness, these opinions are not those of the U.S. DOT which as a matter of policy does not endorse products. With that I think we are finished with the T-3 webinar. I would like to thank our presenters for some terrific presentations and the endurance for answering all of those questions.

back to top