So recently, there has been some buzz around issues with the Wi-Fi on the train. Specifically people have been complaining about how slow the connection has been and how it can cut out at certain points on the route and how the system overall was a piece of __insert your own expletive here___. Being a sensitive soul and one that doesn’t like to see a good thing turn sour I did a little digging to figure out what the extent of the problem is and what can be done about it.
So if you remember a while back, we had plans to update our main website over at www.capmetro.org. The current page format and structure is a little mature in web-years and we had identified a number of new features (with your help) that would be a lot more useful than what we have now. If you are the type of person to wonder what happened to plans of yore, you may be interested in learning what became of the website update. If you aren’t, then feel free to read someplace else. I promise I won’t be offended. Continue reading “Whatever Happened to That Web Update?”
Sorry for the long delay in reporting from the technology fronts. We have been extremely busy getting ready for the rail startup in just a few weeks (working on all of the technology at the stations and behind the scenes that you can’t see) and in pushing the Automatic Vehicle Location project forward to fruition (as I started to talk about here).
One of the exciting things we are working on is enhancing the trip planner and website to provide better information about where the vehicle you care about is and when it will arrive. The latter problem while not trivial is actually the easier of the two to report on. In fact we have recently enhanced our website to give predicted arrival times for the next 3 buses at any given stop (link). Please take a look at it and let us know what you think. Currently this information is based on scheduled arrival time, but as we turn on the Automatic Vehicle Location system we will start replacing scheduled time with estimated time (based on the present location of the vehicle and its latest speed).
The trickier bit is if we should, and how we should display the real-time location of our vehicles once the AVL system is installed. The advantage to displaying the latest location of all of our fixed route vehicles is that individual riders can figure out which vehicle will best meet their needs and it greatly improves the transparency of the system. The down side to displaying this information is that people will count on the data being precise and as everyone should be aware of by now, technology is not always as accurate as we would like (think airplane arrival times or medical billing :-)). What we don’t want to happen is to put information out there that a specific bus is 3 blocks away when it is really 1 block away. People will act on the information in front of them and may miss a bus they wanted. This is not what we want to accomplish with AVL.
I think the key is to display the information in a way that quickly indicates how precise and how reliable it really is. All AVL systems have to pick a frequency of vehicle location updates. For bandwidth and communication cost reasons it is impossible to query the bus and train vehicle every second to know where it is. Practically there is little value in querying a bus every second for its location when it is moving at 5 miles per hour. Conversely it is bad to query a train at 5 minute intervals when it is moving at 60 miles per hour (the stated location will be up to 5 miles away from the true location of the train). For this reason our system will attempt to balance the frequency with the velocity of the vehicles and find a happy medium. But as with all things used by many people, it will not be possible to please everyone with the compromise we reach.
Given this challenge of frequency and real-time accuracy we are left with the issue of how to display the information in a meaningful and non-misleading manner. I pose this challenge to the Austin community as I have yet to find any transit agencies with AVL systems that seem to have solved this conundrum perfectly. For your consideration, here are some of the agencies we have found with AVL that are attempting to visually display the most recent location of their bus fleets. You be the judge and let us know what you think works best.
King County Washington Note: Shows last time the vehicle was querried
Chicago Transit Authority Note: Nice display of Google base map and option to pick routes
Next Bus Note: This is a private company that integrates the approach for many agencies
There may be others, and we would love to hear about them, but we would really like to hear your thoughts on this matter.
I will be the first to admit that public transit is less than intuitive. Timetables and routes and zones and fare structures are less than intuitive. And the harder it is to use something (like public transit or your cell phone’s latest features) the less likely people are to use it. In our business of public transit we are aware of the many barriers to using our system on a daily basis (despite what it may seem like from the outside). Working in the technology group there are however only so many of those barriers that we can address directly (much as I would like to I cannot make the buses come more often). But what we can do is try to make the effort to get information about the system as easy as possible. The theory being that as we knock down barriers to using public transit, more people will want to use it. In that vein, I came across a fascinating article about a little town right here in our neighborhood that has gone off and done something useful in a very ingenious way. While the connection to public transit may not be obvious, let me explain…
Out friends to the south in San Antonio did something smart about 10 years back by putting a unique number on every bus stop in their town. With that number, when someone calls into their phone system or when they visit their website they are able to access information specific to their bus stop simply by referencing the unique number at their stop. This makes it way easier to ask for the next bus to arrive at stop #5413 than to have to describe the bus stop (the one just north of 5th and Congress on the um I think it is the west side of the street… hold on a minute and let me ask someone which side of the street we are on…). This type of short hand is very useful for lowing the barrier to bus and train information.
Realizing the advantage of this type of shorthand to reference points of interest within the Capital Metro world we have begun the long process of putting unique bus stop numbers at each of our stops as we roll out new signs (the problem with getting this done quickly is that we have to modify 3100+ bus stops and transit centers with a precise piece of information that can’t be in error). As we start to get this numeric shorthand in place you will see us roll out the new ability to get stop specific information from our web and IVR systems in this way.
What is exciting about the Manor experiment is that they have taken this concept of a simple reference link to a much deeper source of information and they have proven it can be done for a small amount of money and they have shown that a lot more information can be compressed into a relatively small space. The new form of short hand (in their case a QR code) can be used to convey much more information than a simple 4 or 5 digit number. Of course the hurdle now becomes getting people familiar with a new way of accessing information, but as camera phones become more popular this problem may be solved by other people. (To understand what the city of Manor did and to understand this new way of compressing more information please read the associated article here.)
And for those of you that would like to try this out, I have included a QR code jump below to e-mail me your thoughts. (If you need help deciphering this strange beast, read the article above. If that doesn’t work then post a comment on this blog and I will show you how to take advantage of these hieroglyphics.) As always, I would love to hear what you think on this topic and where you think shortcuts like these could best be used in our system.
I have long been meaning to post on our new Automatic Vehicle Location project but given the size and complexity of it, I have been struggling to figure out where to begin. Of course what you don’t start you can’t finish, so here goes….
What is It?
Most commonly called the Capital Metro ITS project, the AVL project has been somewhat misnamed (more about that in a minute). The Automatic Vehicle Location (AVL) project is designed to put GPS (Global Positioning System) devices on all of our fleet so that operations and the public will better know where our vehicles are at any given moment. While a simple concept, the effort required to achieve this is very large and the impact is potentially huge. Over the next few posts I want to explore what the new system will do for everyone and why it is worth the effort.
Why The Funny Name?
When the Automatic Vehicle Location project was first planned and conceived, it was one of the first major introductions of “high-tech” stuff to be rolled out to the public for Capital Metro (not counting Automatic Passenger Counters, and Electronic Fare Boxes). Therefore the term Intelligent Transit Systems was used to describe the project. In general Intelligent Transit Systems are any use of technology that face the public in the area of transportation. So things like smart message signs on the freeway, traffic light pre-emption technology for rapid bus, electronic ticket vending machines at train platforms, video cameras at traffic lights, etc. are all examples of Intelligent Transit Systems. Now that Capital Metro is exploring usage of many new technologies to better the public transit system, it no longer makes sense to call this project the ITS (or Intelligent Transit System) project. So I will attempt to always refer to it as the Automatic Vehicle Location project (AVL for short) to avoid confusion.
What’s in It For Me?
That’s really the question most people want to know the answer to. I hope to go into detail on this over the next 2 -3 posts broken down by the Mode in which we are rolling it out. The project starts with the technology being placed in the MetroAccess (Paratransit) vehicles first where the greatest cost savings and value to the customer can be realized. Once that fleet is done we plan on rolling it out to the rail vehicles, then the Fixed Route system, and finally the rapid bus system when it comes on line.
Next post will be the details in each of these modes.
As promised, not every one of my posts will be about the IVR. It wouldn’t make for very interesting reading no matter how much you love that technology. But while I am on the topic of apologies, there is one more area that I would like to discuss in terms of where we are and where we would like to be: the Capital Metro Web site. As I write, our Web site is being prepared for a number of changes and enhancements that should make it simpler and more useful for the community (and a lot easier for us to update and keep fresh, too).
The first thing our team did was look at other transit agencies to see what they had done that made sense and what mistakes they had made. This, along with many of the things we have wanted to do for a while but did not have the time or skills to implement, led to a list of things we wanted to build into our next generation Web site. While the brainstorming was easy, the prioritization has not been. We want to vet the list of ideas through the public to see which ones ring true and which ones are not worth pursuing (see below).
We are hoping to get a new look and feel to the page (to freshen it up) and to incorporate features that will take advantage of our new bus tracking system that we are rolling out this fall.
So, tell us. What features would be most valuable to you?
Ideas for new web features:
_____ Much improved Customer Comment Request system (inquiry, trouble ticket, etc.) entry, FAQ, and routing to/from the person that needs to respond.
_____ Ability to put performance metrics on the website.
_____ Ability to have the web visitor rate usefulness of content.
_____ A MyCapMetro page whereby personalized information is accessible based on what I’m interested in, such as: common trips (departure and arrival for common routes – such as to work or work to home), next bus for a few highly used (personally selected) routes – perhaps simply scrolling up to three ‘next departure’ times for certain routes at certain locations based off of GPS data, complaints entered and responses to them, and automatic feeds on topics of personal interest from the Web site, such as STS or Rail, or Board Meetings, etc.
_____ Mobile device button/section on first page, like BART’s, that allows phones and PDAs to download schedules.
_____ Aesthetically pleasing tourist section, like NJ Transit.
_____ Service Alert section at top of home page. Service alerts also listed on schedule pages.
_____ Easy online store.
_____ Language conversion like WMATA.
_____ Community Calendar with trip planner pre-encoded to destination.
_____ Trip Planner: enhanced graphic user interface, drop-down menus, links to interactive maps and service interruptions.
_____ Icons on the front page to distinguish modes (rail, express, local, STS, etc.) clicks through to page with pull down menus of routes and include fare information, etc.
_____ Less “What’s Happening” on the front page and more service information.
_____ Front page feature to click through to detours.
_____ Larger font size (currently 8.5 and gray).
_____ Ability to magnify web pages for those that are visually impaired (see http://www.tafn.org.uk/ for example).
_____ More white space on the pages – simpler, less cluttered design.
_____ Easier, more direct link to interactive maps.
_____ Overall ease in finding what you are seeking – no more than three clicks to any page.
_____ Clarity of pages – easier navigation.
_____ ADA: Bobby approved, 508 and W3C certified.
_____ Live chat with customer service.
_____ Front page link for meetings and events.
_____ Front page feature for project Updates (i.e. All Systems Go).
_____ Improved design on schedules page to reduce the long, undefined columns. Redesigned for easier reading and downloading to mobile devices.
_____ Text messages when my bus is late.
_____ How to ride videos.
_____ “Testing area” for new technologies (like Google Labs).
_____ Forum for dialogue with public.
_____ Orbital ITS (next bus and live tracking of vehicles) online.
_____ Web component for paratransit.
_____ Touchless transit pass or smart card that could be recharged online.
I thought I was done with IVR posts. I was wrong. One of the wonderful things about the Internet is the way information gets shared behind the scenes. Since I started writing about our IVR problems I have now been contacted/referenced by two different IVR specialists that say there may be hope for what ails us.
Now I am more optimistic than most but I want to be realistic about what we can and cannot accomplish with our IVR and then I want to set the public’s expectations realistically (don’t worry, what the public has asked for is realistic, the gap right now is on Capital Metro’s side). In order to do that we have to look at all possible options. Including the largest of all which is scrap the system all together and go in a different direction. Some have said this is the way to go. At this point I don’t know if that is the right answer or not. But we won’t know if we don’t look at it. And we will.
To that end, I want to thank all those that have given and are giving us feedback. While no one likes to be told something they have responsibility for stinks, it is unfortunately the best way to get better. And that is what we are going to do. So please keep sending the ideas, and YES, vendors please keep contacting us so that we can eventually get to the best darned Public Transit IVR in the U.S. of A.
Okay, now for something completely different…
So how do we fix the problems with the IVR? In a word, persistence. In a little less brevity we have a few layers to tackle and the plan is to tackle each layer in order. Naturally we want to hit the problems with the biggest impact, but some of the issue will take rebuilding the system from the ground up and some will take new software from the various vendors. Therefore, like other big Goliaths, we tackle the items that will be the quickest to resolve with the biggest payback for the effort.
– Fix the script errors as quickly as we find them. As mentioned before, the number of places that something can go wrong in a system this size is staggering. So while we are looking at the system regularly to find the errors we can, the fastest way to find the issues is to foster feedback from the community that uses the system. When we hear that there is an error in the IVR we have our team take a look and attempt to fix the problem. If it is a simple “misalignment” of the script we try to correct it immediately. If it is a more complex problem, then we often have to refer back to the vendor and are therefore more constrained by their schedules.
– Work on the source of internal data issues. Unfortunately some of the errors in the IVR system are squarely our fault. Bad data in = bad data out. When we forget to include a bus stop, or we misspell one of Austin’s more creative street names, our customers feel the pain at the IVR. Along with the script errors above we will correct these as we become aware of them. But more importantly, our strategy is to work with each of the groups within Capital Metro to make sure everyone puts good data in.
– Add touch tone to every part of the script that we can. This I think is one of the biggest issues with the IVR today. When I first heard about the issues with the speech recognition, my initial reaction was to pull it. But then when I started floating the idea of dumping that feature, I had numerous reactions from people that liked that piece and that had positive interactions with the voice recognition component. In the end, the best solution seems to be ensuring that we put touch tone everywhere we can in the system along side the voice recognition, and give the callers a choice. (Shocking conclusion I know. Choices are good.)
– Rearrange the script to be simpler and easier to use. Include better structure for the voice recognition. By changing the way we approach the voice recognition prompts, by making it easier to “get out of” the IVR cycle, and by better communicating the options that a caller has at any point in the script, we believe that we will have a better product. This change will be tied in with the previous change so that what you get is a more effective tool and will allow us to better pair touch-tone and voice recognition throughout the system.
– Add additional, obvious, and beneficial functionality. This actually will be the hardest change, which is why I saved it for last. The reason new functionality is tricky is because it depends on pulling information out of other systems (sometimes 2 or 3 simultaneously) which requires coordination and boundary discussions. Of course, the easiest changes are the ones where the underlying database or application has the information you want. But it seldom seems to be that easy.
In a nutshell this is going to be a long process to fix the system. I am hoping that as we progress we can stabilize quickly, make some obvious improvements early on, and then continue to deliver a better IVR month after month into the foreseeable future. If you have ideas or suggestions, please let me know.
From my first week at Capital Metro, I heard about the issues facing our IVR system. While there was no shortage of anecdotes, the bigger challenge was figuring out the root cause of the problems. There was no point in putting a band-aid on the IVR when there were structural problems with the system. However the way issues were reported did not immediately reveal the source of the problems. And to further complicate things the IVR is not a single computer running in a closet somewhere. The system is actually made up of multiple phone related servers, the phone system (both on your end and ours), scheduling data stored in one database, and multiple route and reservations data stored in other databases. All told, the IVR system is made up of about a dozen vendors’ products that all have to work together to present the data that Capital Metro puts into it. So when a problem occurs we have to go through the following steps before we can figure out what is causing the problem.
- Understand what problem is happening
- Reproduce the problem consistently
- Trace the problem back to the one or more systems causing the problem
- Get the vendor(s) responsible for that/those system(s) to take ownership of the problem and solve it
The easiest problems to solve are actually content problems (when we enter data wrong or incompletely). They aren’t any easier to find, but at least we are the only ones accountable for these fixes. The issue here again is the size of the system, with the tens of thousands of data elements that have to come together to make it work, we are constantly finding items that need to be fixed.
As I mentioned earlier, many of the complaints about the IVR were anecdotal (certainly valid, but not specific enough for us to reproduce and therefore be able to fix). However, in recent meetings with the riding community (both at ACCESS and Customer Service Advisory Committee group meetings) I’ve captured the specific issues I’ve heard and we’ve been prioritizing them to address the underlying issues. In my next blog I will lay out the priorities and the next steps that we will be undertaking to make sure the IVR gets better. However, in the remainder of this post I am going to list the issues that we have captured so far so that I can fufill the promise I made to the ACCESS and CSAC groups a few weeks ago (I haven’t forgotten):
Given the length of the list and the limitations of this blog, I am going to try to link to the Excel Spreadsheet. If this process doesn’t work, we will regroup and try this a different way.
To understand IVR’s one needs to understand the nature of call centers. IVR’s were invented to try to deal with the explosive growth of inbound calls that companies experience as their customer base grows without increasing call center staffing in a linear fashion. Many calls are common and request simple information that can be easily automated. The theory is that if you can answer these simple questions with an automated system, then you will need fewer people to respond to the types of questions that only a human can handle (and thereby lower the cost associated with customer or business growth). The problem comes when more complex questions are pushed to the IVR’s and/or the callers come to prefer the IVR and thus depend on it to answer more complex questions. The result in either case is a frustrated caller and a potentially lost rider. Like most things in life, the trick is in finding the balance between the purpose for which IVR’s were built, and the need to handle more calls on a daily basis.
For example, when a rider calls in to find out what hours the customer service line is automated, an IVR can and should handle this type of call. But when a rider calls in to find out how to get from Downtown to Highland mall in the shortest time possible, an IVR will not do a good job of handling this question (a lot of human judgment and discretion is required which an IVR just can’t muster). So why do I mention all of this? The simple answer is that the Capital Metro IVR is not currently meeting our rider’s expectations at the level we would like. What I hope to go into over the next few posts is why this is the case, and more importantly, what Capital Metro is doing about the situation.
CIO – Capital Metro