Archived Data User Services (ADUS)

ITS Produce Data The (sensor) data are used for to help take transportation management actions –Traffic control systems –Transit vehicle performance improvements –Traveler information –Commercial vehicle operations improvements –Etc…

So, Did those management actions actually improve travel? Could you have done better? Were the assumptions you used when you built the system valid? Are there other actions you could take?

To Answer Those Questions Collect, store, and use the ITS data you generate to –Understand the performance of the transportation system –Understand the performance of your management systems –Feed the analysis efforts that examine ways to improve the transportation system

ADUS Management of the transportation system can not be done without knowledge of its performance You can’t manage what you can’t measure

What you measure gets managed

ADUS We are now beginning to store the data we already collect The next step will be to actually use the data we store to improve management of the system

ADUS The difficulty comes not from collecting ITS data but from storing the data and making the stored data useful

ADUS Data storage and use requires an archive system –Data collection –Data transfer to the archive –Data storage (aggregation?) –Analysis and reporting Quality assurance Integration with other data sources –Database management

ADUS ITS systems produce data – already - often continuous data Large volumes of data are: –Difficult to handle –Expensive to store and manipulate –Not something DOTs are used to dealing with

ADUS The collection of data for data’s sake is a waste of money Collecting data is only a good idea, if you intend to use it

ADUS To maintain a useful archive (they’re expensive), the key is: Does the data you collect help you run the system more effectively? –Decision support –Performance reporting

ADUS Data can be used for –Operations planning –Maintenance planning –Safety analysis –Facility performance monitoring –Policy analyses –Congestion monitoring –Systems planning –Environmental analysis

But are the data actually used for those purposes?

ATMS – Types of Data Available Vehicle volumes Vehicle speeds Travel time measurements Origin / destination patterns (from tolls) Incident locations and durations A variety of more specialized data items –Volumes by various vehicle classifications

APTS – Types of Data Available Transit ridership Transit vehicle performance Vehicle maintenance performance Signal system performance

ITS Data Continuous data allows measurement of transportation system performance variability Reliability is a key to management of the transportation system

ITS Data Are Different Allows more detailed analysis Requires creativity

Why Creativity? ITS often provide different (better) measures of performance than traditionally used

An Example of Using an Archive Freeway Performance Data Used for Policy Analysis (Are our policies doing any good?)

Traditional Performance Measures Do not describe the complexity of what is happening on the roadway Are not easily understood by most decision makers and/or the public

Traditional Performance Measures Reported measures traditionally are: –v / c ratios –Usually based on limited data, and a poor mechanism for showing changing conditions during the day –LOS –Hard for non-technical people to understand –Based on limited data –Travel time and delay –Often based on very limited sample, or some very imperfect calculations

Improved Performance Measures This allows you to illustrate some very important trends to a general audience in non-technical terms

Improved Performance Measures But volume alone does not tell someone whether the facility is working effectively Color coding speed information on top of volume data yields a more descriptive picture of performance

Improved Performance Measures Unfortunately, averaging speeds over many days often hides the fact of how often a facility is congested

Improved Performance Measures We display that information based on the percentage of time a facility falls below a designated speed

Freq. Of LOS F (stop and go)

Improved Performance Measures This can show you what happens when a major change in freeway operations takes place In this case a switch from outside to inside HOV lane operation It also illustrates other trends

Improved Performance Measures But freeway performance varies significantly from location to location

The Effect of Starting AM Ramp Metering on SR :30 AM7:30 AM9:30 AM11:30 AM1:30 PM3:30 PM5:30 PM7:30 PM Time of Day Frequency of Congestion Hourly Volume Per Lane (vplph) 2001 Congestion 1999 Congestion 2001 Volume 1999 Volume 170 veh / hr / ln improvement LOS F occurs 1 day per week less often

Improved Performance Measures This means that it is important to –Choose your location carefully –Report on multiple locations –Report on geographic variation So we developed a geographic illustration of traffic performance

Time PMAM Time PMAM Montlake Blvd. 520 Arboretum 92nd Ave. B’vue. Way 148th Ave. NE 60th St. 84th Ave. Lk. Wash. Uncongested, near speed limit Restricted movement but near speed limit More heavily congested, mph Extremely congested, unstable flow Westbound SR 520 Traffic Profile General Purpose Lanes 1997 Weekday Average Eastbound

Improved Performance Measures Geographic illustration can be –Volume –Speed –Lane occupancy –Other

Improved Performance Measures The basic geographic by time of day matrix allows us to compute travel times starting every five minutes, 365 days per year These travel times can then be summarized

Improved Performance Measures Travel time summaries include –Average travel time for a given O/D pair by time of day (every five minutes) –90 th percentile travel time –Frequency with which a given average speed is not achieved

Improved Performance Measures Travel times can also be compared –Against standards –For HOV versus SOV

Improved Performance Measures When truck volume and weight data become available for freeways, these same matrices (and some assumptions) can be used to compute: –Truck hours of delay –Truck miles of delay –Ton-miles of delay –Value of freight delay

Improved Performance Measures Each time we use our new tools to answer a question, we develop new ways to display that information The goal is to make that information –Easier to understand –More accurate of “real life”