USF Libraries
USF Digital Collections

Task order #3 : evaluation of the traffic 2000 : pezo elctric teadle


Material Information

Task order #3 : evaluation of the traffic 2000 : pezo elctric teadle
Physical Description:
Florida Dept. of Transportation
University of South Florida. Center for Urban Transportation Research
Center for Urban Transportation Research (CUTR)
Place of Publication:
Tampa, Fla
Publication Date:


Subjects / Keywords:
Traffic flow--Measurement   ( lcsh )
Piezoelectric devices   ( lcsh )
letter   ( marcgt )

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 37870267
usfldc doi - C01-00045
usfldc handle - c1.45
System ID:

This item is only available as the following downloads:

Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader ntm 22 Ka 4500
controlfield tag 008 s flunnn| ||||ineng
datafield ind1 8 ind2 024
subfield code a C01-00045
(OCoLC) 37870267
2 110
Florida Dept. of Transportation.
0 245
Task order #3 : evaluation of the traffic 2000 : pezo elctric teadle
Tampa, Fla
b Center for Urban Transportation Research (CUTR)
c 1997 July
Traffic flow--Measurement
Piezoelectric devices
University of South Florida. Center for Urban Transportation Research.
1 773
t Center for Urban Transportation Research Publications [USF].
4 856


Final Evaluation Report Task Order #3 Evaluation of the Traffic 2000 Piezo-Electric Treadle prepared by Center for Urban Transportation Research College of Engineering-University of Soutb.Fiorida July 1997 The opinions, findings, and recommendations expressed in this report are those of the Center for Urban Transportation Research and the University of South Florida and not necessarily those of the Florida Department of Transportation. 11is report, serving as the final report for Task Order #3, has been prepared in cooperation with the Florida Department ofTransportalion-Office of Toll Operations, in partial fulfillment of (Electronic Toll and Traffic Management Retainer Services) No. B-A900, State Job. Nos. 97000-1322 and 99900-5000, Journal Trans. 4920265500648900700, and CUTRAccount No. 21-17-248-L.O Mark Burris, Research Associate Michael C. Pietrzyk, ITS Program Manager RamaKrishna Apparaju, Graduate Research Assistant


Final Evalua tion Report Task Order #3 Eva lu a tion of the Traffic 2000 Piezo-E l ect ri c Treadle prepared b y Center for Urban Trans portation Research College of Eng in ee ri ng Unive r s ity of South F l orida July 1997 The opinions, findings, and recomm e n datio n s expressed In this report are those of th Center for Urban Transportation Research and the Universily of South Florida and not n ecessarily those of the F l orida Department ofTransportaflon This report, serving as the final report for Task Order 113, has been prepared In cooperafl o n wlfh the Florida Department ofTransportafion OJ!ice of 1'oll Ope rations, In partial fulfillment of (El ectronic 1'oll and Traffic Management Retain e r Services) Co ntract No. B-A900, State Job. Nos. 97000-1322 and 99900-5000, Journal Trans. 4920265500648900 700, andCUTRAccount No. 21-17-248-L.O. Mark Burris, &search Associate Michael C. Pletrry/c, ITS P rogram Manager RamaKrishna Apparaju, Graduate Research Assistant


Executive Summary Florida's Turnpike is currently using TRMI pressure sens i tive axle detectors or treadles to detect vehicle axles. These treadles consist of two, three, or four parallel metal strips held apart by regularly -placed spacers. As the wheels of the vehicles pass over the treadles, the strips act as electrical swit c hes and generate a signal by which each axle is detec ted. As part of the SunPass procurement, F lorida's Turnpike is installing a different type of treadle, the Traffic 2000 piezo electric treadle, in its ETC dedicated and express lanes. To save money and for uniformity across all lanes, the Turnpike would like to inst all the Traffic 2000 treadle in all its lanes. It was found that few companies besi de s Traffic 2000 p roduce this sort of devic e. However, many corporations are involved in the deve lop men t of conventional axle detecto rs. The primary obj ective of this report is to evaluate Traffic 2000 piezo-electric treadles, while other trea d les have been examined for comparative purposes. Althoug h the piezo-electric treadles have been fo u nd suitable fo r hig h speed (greater than 5 mph) app lic a t ions they have traditionally been unable to accurately detect axles in low speed (less than 5 mph ) conditions like those at conventional toll plazas. T r affic 2000 may have solved this problem using new softw are However, dete rmining the effectiv eness of these treadles is difficult because the results are based on only a two-day test on the Dallas North Tollway. Similarly, there are several adverse conditions in which thi s treadle has not been tested with slow-moving traffic. With only two similar systems o p erating in the United States, fo recasting the p ossible maintenance requirements and operatio n al benefits is difficult. However, the design of the device is such that it should last longer than the treadles used now, it will be very easy to replace, the initial costs are lower and the installation easier than the current treadles. Due to the limited evidence available on the accuracy of these piezo -e l ectric treadles at low speeds, further testing is essential to determine their accuracy, durability, and r elia bility in the low spee d ranges. It is recommended that sample piezo-electric tre adles be obtained from Traffic 2000 for independent testing and eva lua ti on. They shou ld be tested comprehensively to work unde r different conditions like conditions of slow-moving and fast-moving traffic, adverse weather conditions, human interferences with the device s, and vehicles traveling at an angle over the treadles. Evaluation of the Traffic 2000 Piezo-Electric Treadle


Table of Contents 1.0 Introduction ........... a 1 2. 0 Traffic 2000 Piezo-Electric Treadle ...................................... .. 2 2.1 What is the Traffic 2000 Pie zo-E lectric Treadle? ....................... 2 2.2 How Does the Traffic 2000 Pi ezo-E lectric Treadle Work? ............... ... 2 2.3 Market Penetration of the Traffi c 2000 Piezo-Electric Treadle .............. 4 2.4 How is the Traffi c 2000 Pie zo-Elec tric Treadle Installed? .................. 5 2.5 What is the Output fro m the Traffic 2000 Piezo-Electric Treadle? ............ 7 2.6 Maintenance Requirements for Traffic 2000 Piezo-E lectric Treadle .......... 7 2.7 Possible Problem s with Traffic 2000 Piezo-Electric Treadle ................ 9 2.7.1 Slow-Moving Vehicles ....................................... 9 2.7.2 Fast-Moving Vehicles with Close ly Spaced Axles ............... 11 2.7.3 Vehicles Traveling over the Treadles at an Angle ......... ......... 12 2. 7.4 Severe Rainfall ..................... . ..................... 14 2.7.5 Tunnels Beneath the Treadles ................................. 14 2 7 6 Cross Talk from Adja cen t Lanes and Other Sources ................ 14 2.7.7 Piezo-Electri c Cable ........................ ................. 14 2.7.8 Vehicles Backing over a Treadle ............................... 15 2.7.9 Software Uniformity ........................................ 1 5 2. 8 Costs of the Traffic 2000 Piezo-Electric Treadle ......................... 16 3.0 Treadles from other Manufacturers ......................................... 17 3 .I Treadles from TRMI. .. .. . ...................................... 17 3.1.1 Overvie\v ................................................. 17 3 .1.2 Installation Requirements ................ .................... 18 Ev al uation of the Traffic 2000 PiezoElectric Treadle


3.1.3 Warranty .............................................. . 18 3. 1.4 Maintenance Requirements .................................. 18 3. 1.5 Market Penetrat ion in North America .......................... 19 3.1.6 Costs ..................................................... 19 3.2 Treadles from International Road Dynamics ........................... 19 3.2.1 Overview ................................................. 19 3.2.2 Installation Requirements ................................ 21 3.2.3 Warranty ................................. ................ 21 3.2.4 Maintenance Requirements ..... ............................. 22 3.2.5 Market Penetration in North America ..... ." .................... 22 3.2.6 Costs .................................................... 22 3.3 Treadle s from AMP .............................................. 22 3.3.1 Overview ................................ ............... 22 3 3.2 Installation Requirements ................................... 23 3.3.3 Warranty .............................. .................. 24 3.3.4 Maintenance Requirements ................... ....... ........ 24 3.3.5 Market Penetration in North America .......................... 24 3.3.6 Costs ....................... ................. ............ 24 4.0 Conclusions and Recommendations ................... ................ .... 25 Bibliography .......................................................... 27 Appendix ...................................... ..................... 28 Evaluation of the Traffic 2000 Piezo-Electric Treadle


L is t of Tables Table I: Characteristics of the Traffic 2000 Pie2o-Electric Cable ...................... 4 Table 2: Comparison of Different Treadles ................................ ... . L ist of Figures Figure I: Traffic 2000 Piezo-Electric Treadle ....................... .............. 3 Figure 2: Output from Traffic 2000 Piezo-Electric Treadle ...... . .................... 8 Figure 3: Axles Crossing the Treadles ................ ............................. 12 Figure 4 : Tire Angle ........................................................... 13 Figure 5: Piezo Ceramic Element in Doub le Resin Encaseme. nt ....................... 25 Evaluatio n of the Traffic 2000 Piezo-Electric Treadle


1.0 I ntroductio n Vehicl e detectors are used to count, classify, and control traffic One form of vehicle detection technology, axle sensors, are used singly to monitor axle passages in pairs to measur e vehicle speed and can also be used in combination with vehicle detectors such as inductive loops for classification and weighing purposes. There are several axle sensor technologie s in use today including pneumatic tub es, piezo-electric axle sensors, resi stive axle sens ors, tribo-electric cables, and capacitive sensors. This srudy examines in detail the piezo-electric treadles developed by T raffic 2000 and several c

2.0 Traffic 2000 Piezo-Electric Treadle 2.1 What is the Traffic 2000 Piezo-Electric Treadle? The Traffic 2000 piezo-electric treadle is a device built to determine the presence of a vehicle's axle. It consists of a piezo-electric cable, a treadle a housing, a co-axial signal cable, and an interface board. It has been installed successfully at many loc ations and is counting axles in many high sp eed (5 mph or higher) applications throughout the world for seven years. Until now, the Traffic 2000 piezo-electric treadles (and also other piezo-electric treadles) have not worked well in low speed applications. Software recently developed at Traffic 2000 has the potential to overcome the problems with low speed applications. The device invented by Glyn Roberts, founder of Traffic 2000, originated as a pedestrian/bicycle detection device and was modified to count mo to r vehicle axles accurately. Traffic 2000 is a relatively small company based in Surrey, England and has seven full time employees and up to ten part time workers. The company often hires Mr. Tony Lambert, an independent electronics expert in England to perform modifications to the software and circuitry of its devices. 2.2 How Does the Traffic 2000 Piezo-Electric Treadle Work? The piezo-electric cable embedded in the tread le (see Figure I) produces different electrical signals when struck or has pressure applied. In the case of the Traffic 2000 treadle, most of this signal is derived from the distortion and stretching of the piezo-electric cable; a small amount of the signal is derived from the compression of the cab l e itself. When this signal is analyzed, the Traffic 2000 software and interface board can determine if an axle has passed over the sensor. Evaluation of the Traffic 2000 Piezo-Electric Treadle 2


Piezo-electric Cable T 30 mm j_ Replaceable Treadle EPDM Extrusion Locking Screw 0 ----Housing Figure 1: Traffic 2000 Piezo-Electric Treadle The replaceable treadle is made from a high performance EPDM extrusion, impervious to attack from acids, a lkalis salts, and ultraviolet rays. Table I lists cable specifications. The piezo-electric cable is attached to a co-axial signal cable within the treadle. This co-axial cable carries the signals to the T2000.57 interface, which is a dual channel piezo-signal conditioning circuit mounted on a printed circuit board. The signa l processing software on this interface board was expressly designed to provide a versa tile link between the Traffic 2000 treadles and th e classification systein used in the toll plaza. To date, only three of these boards exist, and the creation of them is done entirely by hand = Evaluation of the Traffic 2000 Piezo-Electric Treadle 3


by Tony Lam9ert. Tills is an expens ive and t ime-consuming process and makes obtaining a spare board very difficult. However, once Traffic 2000 receives orders for a large number of these boards, they can be mass produced. Recently To liT ex placed an order for200 of the boards, and CUTR may be able to obtain one for use in testing. The o u ter housing is made from a two-part P.Olyurethane that has a hardne ss of 60 shoreD degrees and held in place using a special sealant (see Section 3). (Information from Traffic-2000 brochures) Table 1: Characteristics of the Traffic 2000 Piezo -Electric Cable . v4? _,). __ -. -_---. i"* ; f ,:?,:, .... . ....... .. :,. .-. S" Y ., ........ . '<' , -, .. >" "':V"*"*'&.,...;,(::)O('j$ .. j.,M;, ; '*' <)< # -.,.;. _.! .' Piezo-electric coefficient 25pCIN at23 oc Temperature coefficient 0.5% per degree C Capacitance 750 pF/m Insulation resistance >to n Temperature range -40C to +70C 2.3 Market Penetration of the Traffic 2000 Piezo-Electric Treadle The Traffic 2000 piezo-electric treadles are in wide use around the world for high speed applications. More than 25,000 treadles are in use worldwide. However, use of these devices in slow-speed applications includi ng toll plazas are extremely limited. There is on e test sight on the Dallas North Tollway (one lane), another in the Mountain Creek Toll Plaza (one lane), a test sight on the Pennsylvania Turnpike (one lane), two toll lanes in France, two toll lanes in Brazil, and at least three lanes in Venezuela. According to Traffic 2000, the toll authority in Brazil is so pleased with the device it has attempted to copy it illegally. Evaluation of the T raffic 2000 Piezo-Electric Treadle 4


2.4 How i s the Traffic 2{100 Piexo-Eiectric Treadle I nstalled? Exact location of the treadles is the responsibility of the systems integrator. The systems integrator determines how best to use these devices in the lane; the devices themselves simply detect the passage of an axle and report it to the lane controller. To facilitate installation of the Traffic 2000 tread l es in an existing lane already equipped with a treadle (AMP, IRD or TRMI), the manufacturer suggests that the old treadle be removed and the area be refilled with concrete. Two 2-inch wide cuts are made in the pavement exactly 50 centimeters apart. Carefully measuring the distance between the two cuts (and eventually the two sensors) is important because the speed of the vehicle can be detennined by knowing the time it takes an axle to pass over one sensor and then the other Current software assumes a 50-centimeter distance between sensors, making it unlikely that humans will step on the treadles in such a way as to cause an accidental axle count. A distance other than 50 centimeters can be used, but if the agency also wants to determine speed using these treadles, then Traffic 2000 would have to program the software for this alternate distance. The outer housing (made from a two-part polyurethane) (see Figu re 1) is then placed into the sawcuts flush with the road bed or just slightly protruding. The side of the housing with the mounting screws is located downstream of traffic. This housing should not be left in the road without a treadle or dummy in it because heavy traffic could cause the s ides o f the housing to cave in. The housing i s then bonded to the road using a special sealant made in the United Kingdom This sealant is classified as a hazardous chemical because, when heated, it p roduces hazardous fumes that should not be taken internally. However, this should not be a problem during installation if installers wear gloves and exercis e caution, and remain in open air Transport of the chemical between continents is difficult and time-consuming; however, Trafftc 2000, To!ITex, and their supplier are working on stocking supplies in the U.S. Evaluation of the Traffic 2000 Piezo-Electric Treadle 5


. Three solutions t o the problems presented by the sealant are being examined The ftrst is to use the sealant despite its hazardous material classification; this would entail a bette. understanding o f the chemica l t o ensure it is not hazardous when used as designed. The second solution is to try various sealants produced in America, such as Appleton's APELCO Sealing Cement and Filler or one of the many approved by Ca!Trans. However, Traffic 2000 has not been abl e to find a suitable replacement product made in Nort h America. F indin g a proper replacement would also e ntail extensive testing to ensure it meets both Traffic 2000's and tbe transportation department's specifications. The UK sealant has successfully und ergone 18 months of testing by the United Kingdom Ministry of Transport. The sealant bonds and h ardens very quickly and does not allow the housing to move or settle. Attempts were made to use a product called E-Bond at the demonstration site on the PeMsylvania Turnpike. However, it allowed too much movement and settling and had to be removed. The third solution, which is now being pursued, is to have the manufacturer of the current sea lant develop a non-toxi c substitute. Once the outer housing is in place, th e replaceable treadle can then be fitted inside the housing. The treadle consists of an EPDM extruded plast\c housing surround ing a piezo -electric cable that sends the signal to the Traffic 2000 software interface. The treadle is gently knocked into th e housing with a rubber mallet or, alternatively, a piece of timber one inch thick is placed over th e t readle and an ordinary hammer is used_ This entire installation process (two treadles, cable connections, and the circuit board) could take a novice approximately five to six hours. An experienced insta llation technician can perform the entire process in just over two hours. There is much l ess work involved in th e installation of these devices than th e TRMJ or International Road Dynamics (IRD) devices. Furthermore th ese tr eadles do not require expens ive metal frames or the building of a water drain. Also, when one of these devices has to be replaced, only the inside housing and cable need to be re placed. To do this, the l ocking screws are removed and the treadl e i s pulled out of the housing. The signal cable is then Evaluation of the Traffic 2000 Piezo-Electric Treadle 6


pulled through the conduit with a pull -thr ough cord. The pull-through cord is attached to the signal cable on the new treadle and pulled back through the conduit. The inside of the housing is cleaned with a rag and the new treadle is then in serted in the housing and the locking screws replaced. A new BNC is cri.mpedonto the end of the signal cable and connected. This entire operation can take Jess than half an hour. This housing and cable are hammered into the outer housing and held in place using locking screws 2.5 What is the Output from tbe Traffic 2000 Piezo-Electric Treadle? The device measures the impact and vibrations caused w hen something hits the treadle or is on the treadle. The software interface receives a signal from the piezo-electric cable indicaiing the distortio n stretching and compression in the cable. The software will then analyze and process this signal to detennine i f an axle has passed over the sensors or not. Often, and in the case of SunPass o n the Florida s Turnpike this is the only output required from the tr e adle An example of the unprocessed data r e ceived from the senso rs can be seen in Channel s I, 2, and 3 in Figure 2. The channels are located on theY axis and the X axis represents t ime in 20 millisecond increments. Channel 4 data represents the processed signal and clearly shows two tir es crossing one treadle I 00 milliseconds apart. The data that are passed f ro m the Traffic 2000 software controUer to the lane controller can be modified to w h atever is needed, including mimicking the signal sent by TRMI treadles used now. 2.6 Maintenance Requirements for Traffic 2000 Piezo-Electric Treadle This device has generally been used for counting axles of traffic moving at highway speeds In these situations, the device is. expected to last for at least 25 000,000 operations (i. e. axles) In a heavily traveled lane (for example, 15,000 vehicles or 35,000 axles per day), these devices would be expected to last almost two years There is at least one treadle in operation in the U .K. that has logged more than 150,000,000 axles. The United Kingdom Ministry ofTransport generally replaces the treadles on their motorways every I 8 months whether or n ot they need replacement. Due t o the high cost of setting up equipment to close Janes and sending crews out to replace treadles it was ,. . .. wn ... r . . ......... -rx == .. yn "=p ,, "tt!?":!t!!'MSW'2Z'TZT r X W t::t!O::O:::t'!l'::: ... T'r:e"TT'XT"'" '11' Evaluation of the Traffic 2000 PiezoElectric Treadle 7


Tek 1-i(e]!l! Single seq soo S/s .. ., . . . . . . . . . : . . . . . 0 . . . . . . 200mV Ch4 5.00 V Figure 2 : Output from Traffic 2000 Pie:to-Eicc tric Treadle Evaluation of the Traffic 2000 Piezo-Electric Treadle 8


found that replacing all treadles in one location, one lane at a time, during a single day is better than simply r eplacing a single treadle whenever it failed Operating in a free-flow environment and in stop-and-go conditions is very different. The stop-and go conditions of the toll plaza environment may prove to wear out the treadles faster. However, the few Traffic 2000 treadles installed in toll plazas have lasted a long time (including Mountain Creek Toll Bridge in Texas where the treadles have been in place for approximately 30 months). Predicting the life expectancy of these devices in the toll plaza environment is difficult Traffic 2000 is offering a warranty of one year or 25,000,000 axles. This works out t o more than 25,000 vehicles per day, which is so high for a toll lane the one year time limit will most likely occur before this many axles are counted. The most significant maintenance issue is the ease with which the treadles can be taken out of the large housing and replaced. This is a very short operation that should take one maintenance person less than half an hour to complete. This is substantially less time and effort than replacing a traditiona l TRMI or IRD treadle. 2.7 Possib le Errors with Traffic 2000 Piezo-Electric Treadle 2.7.1 Slow-Moving Vehicles The primary drawback to using piezo-electri c treadles in a toll plaza environment is their difficulty in accurately counting the axles of slow-moving Oess than 5 mph) vehi cles. This problem appears to have been overcome with the latest software for the Traffic 2000 pie20-electric treadle. In a two day test at the Dallas North Tollway, it was demonstrated that the devices could accurately detect and count axles of extremely slow-moving or even stopped vehicles However, a two-day test is not sufficien t to prove that these devices can accurately count axles of vehicles. A separate, informal test has been conducted on the Pennsylvania Turnpike since Ju. ne 1996. TransCore, an ETC integrator, is running the test independent of the Turnpike and the Turnpike Evaluation of the Traffic 2000 Piezo-Electric Treadle 9


does not wish to be associated with the test This test is in Pennsylvania because TransCore's h eadquarters is located there. This installation did not include the latest software (developed in May 1997 and used in Dallas), but it was used to help develop this new software. The latest software is now being installed and limited testing will continue under the direction ofTransCore TransCore felt that not attempting to be part of this t es t would be best for CUTR and to hold our own tests in our area for two reasons : 1) T he Pennsylv ania Turnpike has been a rel uc t ant partner to this testing and any involvement by CUTR tn ay cause the Turnpike to stop the testing, and 2) TransCore is unsure of when and what tests they will run and have more immediate concerns than testing this device for long term performance TransCore .is examining the use of this device in other toll piau applications around the world. Gil McGlaughlin with TransCore has been involved in this test and indicated that the results to date have been the best for any piezo-electric treadle attempting to count axles of slow-moving vehicles. However, they have been far from perfect, and occasionally the software would require some minor adjustments. The need for these adjustments seemed to occur at the same time as the change in seasons, bu t no definitive reason for the problems was found. Mr. McGlaughlin is reluctant t o reveal these test results. He indicated a need for a new t est with video tape verification of counts and strip chart recordings of the piezoe lectri c signal to understan d the abilities of treadles fully. This test site is located at a toll booth where the first treadle is used for pre-classification on traffic that is starting and stopping. The second treadle is used for post-classification on vehicles that are accelerating To date the post-classification treadle has yielded better results (as high as 99.97% according to Mr. McGlaughlin) than the pre-classification one. In fact, the pre-classification one provided such poor results its monitoring was discontinued. The reason for the poor results is that the board originally supplied to TransCore had one sensit ivity input for both channels and this input had t o be adjusted for pre -class ifying or for post-classifying. TransCore tried to adjust this input such that the board and treadles would both post and Evaluation of the Traffic 2000 Piezo-Electric Treadle 1 0


pre-classify. This was impossible because that version of the board and software could only do one or the other. The board and software have been completely redesigned to recognize how fast the vehicle is traveling and the impact of the vehicle to determine what level of sensitivity is required. The software has a primary recognition component that determines the sensitivity at which to set itself for that particular axle. It th e n begins a secondary recognition phase where th e bulk of the data on the axle is collected at the correct sensitivity level. Once the axle passes the treadle the software reverts to the primary recognition phase. The results are such that the system can detect and count axles at any speed from zero to 120+ mph, and the same board can handle one treadle pre-classifying and one treadle post classifying. 2.7.2 Fast-Moving Vehicles with Closely Spaced Axles A test was performed on the Dallas North Tollway to determine if the software could handle fast moving, closely-spaced axles as well as slow-moving vehicles. A loaded, tandem axle dump truck was driven through the toll plaza and over the treadle s seven times at 56 mph. The software successfully identified all three axles on the truck. The truck also passed over the treadle at 20 mph 10 mph and extremely slowly, with the treadles accurately identifying the axles each time. According to Traffic 2000, the software identifies events in one millisecond incre ment or less. With typically four feet between axles and a 6"X 6" tire footprint, this leaves 42 inches between impacts on the treadles (see Figure 3). At 100 mph, a distance of 42 inches is covered in 24 milliseconds, which would allow for the software to registe r a gap between axles easily. However, since treadles register a footprint twice its actual size (reducing the effective distance between tires to 36 inches; see note 2.6.3) and the vibrations caused by heavy truck moving at this speed may cause a false presence, this aspect of the devices should be tested more thoroughly. --=w=:re -==xn r== == .. ..... ... o:u Evaluation of the Traffic 2000 Piezo-Electric Treadle 11


> Treadle s 4 8" '-' ___, Figure 3: Axles Crossing the Treadles 2 7 3 Vehicles Traveling over the Treadles at an Angle People have a t endency to exit to ll plaza l anes at slight angles, most likely because they must move to the far left of t he lan e t o pay the t oll in no n -ETC lanes. If the angle of the vehicle's axle becomes excess i ve, the t rea d l e may r egis ter this as two separat e a xl e hits (see F i gu r e 4). I t was found that vehic les would have to be travel ing at sig nificant angles for doubl e counting to occur. Accordi ng to Traffic 2000, "in the software of the d u al channel interface, the output is equal to twice the footprint. This means when a whee l pass es over the treadle a n d it is t raveling grea ter than 1 mph, then th e system do e s no t t ermin a t e the prese nce until the wh eel has left plus the same time again as it took to pass over." If the vehic l e is trav eli ng at less tha t I mph the footprint is even greater. As ind i c a ted in Figure 4, a typical vehicle would have to be trav eling at a significant angle to hav e o ne o f its a xles double -c ounted. However, testing this i n a toll l ane would be beneficial to ensu re that the o u tpu t is indeed twice the foo t print. If the output were simp l y equal to the footprint of these t ire s, the maximum allowab l e ang les shown in F i gure 4 would be cut in half. ,, ..... ,, ... , ... ,,., ... ... ... ,fWH"!!<' '''m u r:u:Tmmmuywy m r r r .. uy-rrmww ..... ....... I ................... ... . ,,, __ ... ,. ..... ....... ., .. Ttt"tt'Wt=' sy:qE valuation of t he Traffi c 2000 P i ezo E l ect ri c Tread l e 12


Tires X Figure 4: Tire Angle 6" x= Distance between wheels 12" effective tire footprint 9= maximum angle X a 55 12.3 10'Lane 65 10.5 75 9.1 Evaluation of the Traffic 2000 PiezoElectric Treadle 13


2.7.4 Severe R ainfall T r affic 2000 was asked if the signal received by the piezo-electric cable would change significantly if th e re were an inch of rain over the treadle 'Thls had not been tested, but it is believed thi s would result in part of the load of the ve hicl e being distributed through that one inch of water . This would th en produce an exaggerated footprint, but not affec t the axle counts. 2 7.5 Tunnels Beneath the Treadles Some toll plazas have tunnels runnin g beneath the toll booths and under the treadle locations Traffic 2000 was asked if these tunnels would affect the signals rece ived by the treadles, it was indicated that, since these tunnels are built to withstand traffic tra veling over them, there would be no effect on the treadle from Jhe tunnel underneath 2.7.6 Cross Talk from Adjacent Lanes and other Sources The, Traffic 2000 trea d le is extrem e ly sensitive and can detect vibrations f rom vehicles traveling in l anes adjacen t to the treadles. However th e software can easily filter out thes e vibration signa l s a l o n g with other (for examp le the slammin g of car doors or the dropping of an obje ct) n on axle signa l s. This has b een shown in the Penn sylvania t es ting A human can mimic a v ehi cle crossing the two treadles and cause the software t o count an axle erro neou sly. Howev e r, someone accidentally stepping on the treadles would not cause this t o happen The person would have to be deliberately attempting to fool the system The effort and difficulty in producing this incorrect signal should be examined further to determine the potential difficulties that may arise from this situation. 2.7.7 Piezo -Electric Cable As with any equipmen t obtaining t op quality components for the entire system is cri tical. Currently, Traffic 2000 is havin g diffi culty with pic:ro cable supplied by NTK in Japan T h e batch of cable from NTK was proven to be offaulty manufacture Traffic 2000 is now usin g either Philips Evaluation of the Traffic 2000 Pie:ro-Electric Treadle 14


class 2 or AMP piezo cables, both of which proven reliable and durable. They do not use P hilips class l because it is designed for weigh in motion sensors and the sensitivity levels offered by this cable are are not required for toll treadles. 2.7.8 Vehicles Backing over a Treadle For the sys tem to register the passage of an axle, the axle must pass over the first and then the second treadle. I f the axle does not pass over both treadles, then the axle is not counted. In thi s roaMer, any extra coun ts from vehicles ro lling over tl1e first treadl e and then backing over it will be ignored. Similarly, if the front axle of a vehicle travels over both treadles, the n backs up over the second treadle, then goes forward again over the second treadle, this axle will be counted only once because it only went over the first treadle once . Only in certain extreme situations would an e xtra axle to be counted. For example, if the front axle of the vehicle travels over both t readles and the back axel travels over only the first treadle, then the vehicle backs, and all the way out of the lane, one axle will b e counted More investig ation of how th e integra ted syste m can handle vehicle s backing u p is required, but this scenario is extreme and unlikely to occur often. 2.7.9 Sofhvarc Uniformi ty Will one board and one software program work in all lanes? Will the slight physical differences in lanes the difference in average speed of the vehicles in ETC and non-ETC lanes, or the difference in average weight of the vehicles in a heavy vehicle lane cause the need for separate boards or software? During the short test in Dallas, the board and software were able to accurately process and identify axles from a wide variety of vehicles, from fast-moving dump trucks to slow/stopped passenger vehicles. More lon g term testin g is n eeded, but early results are very positi ve. Evaluation of the Traffic 2000 Piezo-Electric T readle 15


2.8 Costs o f the Traffic-2000 P iezo-Electric T r eadle The cos t of the treadle varies depending on the dimensions of the sensors (frafftc 2000 piezo electric treadles are available in five different lengths of 2, 2 V,, 3, 3 V., and 4 meters), type of sensors including s t andard sectio n sensors or narrow section sensors, the type of h ousing (two different types of housing are available), and the sensor cabl e s. According to Traffic 2000 information brochure, a typical two-meter Traffic 2000 piezo electric treadle costs $550, and a four-meter treadle costs $890 The prices of housing are similar for both s t andard and narrow sections, and ranging from $290 (for a 2 meter long section) to $580 (for a 4 meter long sec t ion). Apart from sensors and housing cables are also a significant portion of the cost, around $200 depend i ng upon the length of the cable required. However, the price ment ioned above includes the c o s t of the cable of length ten m eters. Acco r d ing to Florida's Turnpike, the Traffic 2000 piezo-electric tread les will cost around $2000 per lane for everything required to fully install and integrate them. The replacement costs are about $325 per treadle. Evaluation of the Traffic 2000 Piezo-Electric Treadle 16


3.0 Treadles from other Manufacturers This section briefly describes treadles developed by three other companies, namely The Revenue Markets, Inc. (TR.Ml), International Road Dynamics (IRD), and AMP lncmporated. Each description includes a brief overview of the treadles, the technologies used, their abilities, installation requirements, maintenance requirements, proliferation, and approximate cost. Much of the information in this section of the report comes from the treadle manufacture rs themselves. The TRMI and IRD treadles represen t the vast majority of treadles used in to ll plazas in N orth America. 3.1 Treadles from TRMI 3.1.1 Overview TRMl bas been involved in the design and manufacture of two and four contact treadles. The four contact treadles are replaceable and the two-contact treadles are completely sealed and throwaway installations. The four-contact treadle is equ i pp ed w ith four individual strips, and each of these strips is mold ed around two stainless steel rods. The four strips are then sealed individually and inserted into the cavities in the outer shell. The outer shells of these treadles are made \vith a special rubber composition called GRS to resist the impact of sunlight, salt water, and oils TRMI's two contact strips are constructed with spring-hardened steel strips and, unlike four-contact treadles, the two-contact treadles are connected to the ground with a common base plate. As a vehicle passes over the treadle, the stainless steel rods in each strip are compressed, and the two rods in each strip connect. This causes a signal to be sent to electronic interfaces, indi cating an axle. The rods in the four individual strips close in a specific sequence, determining the direction of the vehicles crossing the treadles. The two-contact treadles are surface mounted treadles and do not require deep slots (11/4 inch deep slots are sufficient.) These treadles suit application where the excavation of pavement is difficult. Their lightweight fu lly-sealed, and single cable construction is useful for applications on both high Evaluation of the Traffic 2000 Piezo-Electric Treadle 17


turnover areas and for smaller applications where the extreme durability of the four -c ontact treadles is not needed. However, the biggest disadvantage of two contact treadles is that they can not be repaired. The four-contact treadles provide accuracy in audit control and have a long life. These devices are expected to be operation for more than 10 years without any m ajo r problems, and the components of these devices are also desig ned to be replaceable. These treadles are available in various sizes. Two contact treadles are available in 6, 8, and 10 foot lengths Four-contact treadles can be obtained in two different lengt hs of 8 ft 6 inc hes, and ten inches. 3.1.2 Installation Requirements These treadles are inserted in frames and embedded in the road surface by making cavities in the road bed. The size of the cavity depends upon the type and size of the frames used. Heavy duty frames require 18 inch deep trenches while the light duty frames require only 6 inch deep trenches. The heavy duty frames require deeper trenches than the light duty frames as they are equipped with additional anchors for accurately positioning the frames. These heavy duty frames have larger drainage capacity with drain boxes on either side of the frame. TRMI suggested the use of heavy duty frames for the toll lanes with more than 20 percent of the traffic being truck traffic and light duty frames for normal traffic conditions. The light duty frames do not have any drain boxes. According to the manufacturer, normal traffic cond it ions are not expected to cause any significant hydraulic action, and hence the drainage boxes are not required. The replacement of faulty treadles is simple and can be completed in Jess than an hour. However, new installat io ns require considerable skill and time. According to the manufacturer, it takes six hours for a new installation and 20 to 30 minutes for treadle replacement. 3.1.3 Warranty TRMI has a long history with a proven track record. Some treadles have been used extensively for more than I 0 years without any problems. However, TRMI guaranties their four-contact treadles for a total of six million counts or two years (whichever comes first); the warranty period for two-Evaluation of the Traffic 2000 Piezo-Electric Treadle 18


contact treadles is one y ear o r one million counts, whichever comes fus t. They have some flexibility wit h t his warranty. 3. 1 4 Mai nte nance Requirements According to TRMI, their two-conta c t and f our contact tr eadle s do not requir e any main t enance as th e y are p ermanent installations and will not fail in adverse conditions like snow, rain, floods, e tc., 3.1. 5 Market Penetration inNorth Am e rica TRMJ treadl es are being used ex t ensivel y throughout t he United States. FOOT uses th e two-contact treadles on many highway s in F lorida, including throu ghout the T urn pike system. 3 .1.6 Co s t s The c os t of the TRMI tre a dles varies depending upon the l ength of th e tr eadle. According t o the manufacturer, a 6 -foot treadle co sts $535, an 8 -foot tread l e c osts $575, and a 1 0 -foot one cost about $625. These costs are only for treadle s an d d o not include th e costs of the frames. T h e frames for TRMI t r ea dles are expec ted to cost around $1375 and these costs are negotiable depending u pon th e vol u m e o f the quantity being pur c hased. Howeve r the sources at Florida Turnpike in dica t e d that new inst a llat i ollS with TRMI t readles could cost ar ound $5000-$8000 and their replacemen t costs will be around $5 00. Th ey also indicated that it cos t $ 5000 for them t o repl ace an existing TRMI treadle frame and construct a new frame on the Bee L ine East E xpressway 3.2 Tread l es from Internatio nal R oad Dy n amic s 3. 2.1 Overview In t erna tion a l Road Dynamics, Inc. (IRD), a company based in Canada, i s ano t her lead ing corporatio n in the design and installation of axle d e t e ction systems. It has develope d s eve r a l types of traffic sensors, including treadl es. IRD developed a two-position treadle known as IRD MODEL AS408 DYNAX treadle. These de v i ces are permanent, re s istive axle sens o rs an d are constructed Evaluation of the Traffic 2000 Piezo-Electric Treadle 19


of a resistive mat eria l that is sensitive to pressure. These sensors are in an inverted T cross-section and are enclosed in a semi-rigid material, imperv ious to moisture and roa dway deicing chemicals. The DYNAX sensor is approximately 1 square inch in area. Both tl ie frame and treadle nameplate are constructed in s t eel, painted with red primer. IRD has also developed speed-independent treadles called IRD DYNAXR treadles, and these devices are considered ideal for toll plazas where speed s impose several constraints. IRD treadles are available in several models, including 1 2, 3-, and 4-strip sensor models. These devices are supplied with frames for brand new installations, and they can also retrofit into exis ting frames. They are offered as a conventiona l two-piece units with a treadle insert and road installation frame and as a unit (uni-body) construction. also offers different models ofDYNAXR tread les The IRD DYNAXR model AS41 0 and AS415 treadles are formed from a luminum extrusions with r em ovable bars that hold the sensors in place. This feature allows for the removal and installation of a sensor without the need to remove the tread l e from the roadway. The lengths of AS41 0 and AS415 model treadles are 8 feet and 10 feet, respectively. Both the models arc four-position treadles, and the inst allation frames are made of heavy-gauge steel C-Cbannels. They employ integral pavement anchors to secure them firmly to the cavity that is saw-cut into the roadway. Frames are painted with red primer for protection against corrosion. The body of DYNAX treadles manufactured by IRD, Inc, is made with high strength aluminum alloys to make them durable and rugged. These sensors are also designed to resis t the pressures from traffic and road maintenance equipment. The four-strip DYNAX treadle is believed to be useful for heavy traffic areas and for low-speed areas like toll plazas. These-four strip devices have the capability of count ing the axles accurately, and they also can determine the direction of vehicles better than two-strip and three-strip devices. When a vehicle rolls over th e four-strip DYNAX treadle, each sensor is momentarily activated in tum, and the lRD fou r-C hannel interface board connected with sensors produces the output signals. By using the customer's e l ectronic equipment, Evaluation of the Traffic 2000 Piezo-Electric Treadle 20


these signals would then be transformed to provide information on axle counts and directions. If the customer's electronic equipment could not transform the signals into required information, new equipment may have to be purchased. IRD, Inc, can provide such equipment, and the specialists at this corporation also provide integration services. 3.2.2 Installation Requirements In the case of DYNAX sensors developed by IRD, Inc., the frames should be installed in a level portion of the roadway. The installation ofDYNAX sensors starts with the saw-cutting of the cavity (20 inches wide by four inches deep by the length required) in the pavement to place the frames. The cavity at the cable end should be deepened up to 12 inches over a I 0-inch span to accommodate the electrical box. The AS41 0 model requires I 08 inch-long cavities, and the A.S415 model requires 126 inch long cavities Conduits are installed in trenches leading to the treadle frames. For both the TRMI and IRD treadles there is somedebate about whether a drain is required. Currently, both treadles require a drain This is a time-consuming and expensive process. 3.2.3 Warranty IRD, Inc, warrants the electronic equipment and internal electronic components for one year from date of shipment. However, the warranty period for inroad instruments and sensors is only 90 days from the date of delivery. The warranty covers the manufacture of the equipment including manufacturers' workmanship, material defects, and assembly, but it does not cover physical damages caused by external conditions such as flood, frre, lightning, or accidents. The IRD DYNAX treadle installation frame and treadle body arc warranted for one year. TransCore believes that the life of these treadles will depend on the type and volume of traffic and that weather conditions may not contribute to the failures. IRD treadles installed on Orlando Orange County Expressway Authority (OOCEA) lanes typically last for at least two to two and a half years w i thout any failures. Evaluation of the Traffic 2000 Piezo-Electric Treadle 21


3.2.4 Mainten ance R equirements According to IRD, these treadles do not require any physical maintenance once installed in the treadle frames, even during adverse weather conditions. TransCore conflnns statement Once the IRD treadles are installed, there is no need for any maintenance until their failure, and once they fail they have to be replaced. The IRD treadles installed on 193 lanes of Orlando Orange County Expressway Authority (OOCEA) are working quite sa t isfactorily u nder severe rain and t emperature conditions. 3.2.5 Market Penetration i n North America IRD has an impressive user list for its treadles in the United States. At about 200 treadles have been in use on Orlando Orange County in Florida. In addition to Florida, IRD treadles are being used t hroughout the country, including in California, New Jersey, and Colorado 3.2.6 Costs The cost of IRD treadles varies from model to model and, with all models, the treadle frames are a significant portion of the cost. According to the IRD sources, the AS 410 model treadle s (8 feet long) will cost about $2300 (with interface, sensor, and the insert p late ) and the AS 4 I 5 (I 0 -foot long) model treadle cost approximately $2900. Out of these prices, the sensor itself is expected to cost around $300, and the treadle frames will cost around $600-$700. The bulk of the expense is in the initial installation of the devices, with the cost of the frame at :uound $1000-SISOO. It is also a tirne-ronsuming process, with a new installation requiring the services of two persons for two days. 3.3 Treadles fr om AMP 3.3.1 Overview AMP, Incorporated, has developed piezo film senso rs that can be installed either permanently or temporarily onto the road surface for the collection of traffic data. The Roadtrax BL Traffic sensor is designed for permanent or temporary installation onto the ground and the Roadtrax BLC (Brass Evaluation of the Traffic 2000 Piezo-Electric Treadle 22


Linguini in a Channel) sensor is a permanently installed piezo-clc:ctric axle sensor. The sensors have to be installed into a slot cut in the road, which is then backfilled with an encapsulation material. The Roadtrax BLC sensor utilizes the proven technology of highly compressed piezo electric polymers, and the sensor element is protected by encapsulating it in a sand-epoxy in an aluminum channel. The sensors are available both as a Class I sensor with the big)ler level of uniformity needed for weigh in motion applications and as a Class II sensor for counting and classifying. The sensors developed by AMP, Incorporated, are provided with insulation to resist gasoline, diesel fue l water, road salt, and transmission fluids, and they are also capable of operating at severe temperatures (from -500 to 16000 F). l}lese sensors are highly flexible and therefore can conform to the road profiles while maintaining a uniform distance to the roa d surface. These sensors are also designed to have an excellent signal to noise r atio, with 10: 1 rejection rate due to road bending, adjacent lanes, and bow waves of approaching vehic l es. They are durable enough to withstand normal ins t allation handling and hundreds of millions of equivalent single axle loads (ESAL ) 3.3.2 Installation Requirements The piezo film sensors developed by AMP, Incorporated require 1112 inch wide by 1114 to 1112 inch deep slots to be cut in the p avement. The slots should be approximately 6 inches longer than the sensor. The lead out has to be centered on the slot. Three holes (one each at 6 inches from the end of the sensor and another in the center), that are 3/8 inches in diameter and V. inc hes deep are then drilled in the bottom of the slot on the center line. High pressure water or water and compressed air are used to clean all the foreign matter o u t of the slot and five inches to all sides of the slot. Excess water and debris are then removed with a vacuum cleaner, and slots are dried using torches, torpedo heaters, electric heaters or natural evaporation, depending on the weather conditions. Sensors and signal cables are then be placed in the slots in such a way that the top of the sensor is slig)ltly below the road surface (about 3/16 inch.) Epoxy has to be prepared according to specifications and be Evaluation of the Traffic 2000 Piezo-Electric Treadle 23


poured on one side of the sensor placed in the slot, and finally the lead in the cable would be routed through the slot cut for it. When the epoxy is fully cured, it could b e opened for traffic. 3.3.3 Warranty The warranty structure is flexibl e and is negotiable, depending upon the type of application and clients However, they do provide a one year warranty for both pennanent and temporary treadles. 3 .3.4 Maintenance Requirements The piezo-electric sensors require regular maintenance. The infonnation broch u re from AMP says its sensors should be inspected on a biannual basis, and any cracks in the road or i n the sensor encapsulation should be filled. These a l so require periodic filling of cracks with a low viscosity loop sealant such as Bondo 606 or a low viscosity epoxy such as Global PX 768. Any loose asphalt material should be wire brushed away prior to pouring the sealant material in the cracks. The sensors should also be tested for capacitance and resistance, and the results Jogged in on the data sheet for the sensors. 3.3.5 Market Penetration in North America AMP's brochure indicates that its Roadtrax sensors are being used by a number of states, Canadian provinces, and countries. 3.3.6 Costs The costs of AMP treadles vary, depending on the model (BL and BLC), the dimensions of the treadles, and the volume of the order. Typically, a six-foot BLC class II treadle costs about $310 $370 and a 12-foot BLC class 1 sensor costs about $620-$750. Another type of treadle offered by AMP (BL treadles) is also available in differen t price ranges, from $270 for a 6-foot ueadle to $510 for a 16-foot treadle. Evaluation of the Traffic 2000 Piezo-Electric Treadle 24


Figure 5: Piezo ceramic element in double resin enca sement 4.0 Conclusions and Recommendations The Traffic 2000 piezo-electric treadle has proven itself to be accurate and cost-effective for high speed applications. The device is cheaper and easier to install than the traditional treadles used in toll plaza operations (IRD and TRMI treadles). The device is very easy to replace once i t wears out, and has a long life. T able 2 highlites the information gathered on the treadles explained in this report. Reportedly, the device has overcome its problems of accurately counting axles in a slow speed toll plaza environment, but this has not yet been proven. Before installing this device in low-speed lanes, testing it extensively is important. Principal testing areas should include examining the treadle's ability to work in severe weather testing new sealants, finding what angle vehicles can travel over the treadle without causing a do uble count, determining how difficult is i t for humans to walk on the trea dles and cause an axle count, determining if tunnels beneath the treadles have any affect on the signal, and determining long-term accuracy over nine classifications of traffic, particularly slow-moving vehicles. Evaluation of the Traffic 2000 Piezo-Electric Treadle 25


i ::r. g 0 ...., ....; ., 5. i N 0> Brief Descr i ption Approximate cost nstallation Maintenance/ Rep .air Requ ireme nts f ypical Life Replacement (Ansidcrations 0 0 Table 2: Comparison of Differ ent Treadles Traffic 2000 TRMI IRD 0 AMP Piezo-electric treadle with Two-Contact and Four1, 2-, 3 and 4-strip Roadtrax piezo film Piezo-electric cab les, coContact treadles with Dynax Treadles with treadles. axial signal cables, and special GRS rubber Aluminum extrusions and interfaces. composition. removable bars. Require frames. Require .frames. It cost $250-$600 per Each treadle cost around Complete installation cost $400-$500 per treadle. treadle. Complete $600 and a new around $2200-$3000. The installation will cost installation co u ld cost treadle itself will b e around $2000 p er lane. around $5000. approxima tely $700 The entire installa tion 0 The installat ion requires It is a time-consuming Two to three man hours process takes two to six six hours of labor. process and require the per sensor. hours. services of two people for one to two days. No results are available Do not require any Do not require any Sealing of cracks is regarding the maintenance maintenance. maintenance. required on a bi-annual requirements ofthese basis. treadles. Arc e xpected to last for up Expected to last at least Will last for at least one 40 million axles to 25,000, 000 hits for three years. and a half years without any failures In case of failures, only It takes 20-30 minutes for Consumes very little time Replacement of treadles a inside housing and cable replacement and the (about two hours) for toll lanes takes about half need to be replaced. This replacement cost per replacing and it will cost a day for two people and operation take less than treadle is around $500. approximately $500costs around $500. half an hour and cost The replacem ent of entire $1000. $325. system may cost approximate l y $5000.


Bibliography AMP, Inc. "Roadtrax Traffic Sensors" (brochure). Valley Forge, Pennsylvania, 1997. Barrier Systems, Inc. "A Move to Intelligent Highways" (brochure). Carson City, Neyada, 1997. Bickley, A. and Moore, T. "Axle Sensors Measure Up." Highways Volume 57, 1989:10-11. Electronic Control Measurement, Inc. "Piezo-Electric Sensors" (brochure). Manor, Texas 1997. IRD,Inc. "A world oflntelligent Transportation Systems" (brochur e). Saskatchewan, Canada, 1997. Laylo, H. M "Piezo-Electric Automatic Vehicle Classification System Report #SHRP-OR RD92-01: Oregon DOT, July 1991. Roberts, Glyn. "Mean Machines." Highways, Volume 60, 1992: 29-29 . TRMI,Inc. "Two and Four Contact Treadles." Accord, New York, May 1997. Evaluation of the Traffic 2000 Piezo -E lectric Treadle 27


APPE NDIX A: L is t o f Co n t a ct s .! .:; :r 'fltc'IT""' ; "' l i..-'":' ,,;; .... :::ff ameu .. ..., .. ,,.'lll! P!lUY:. ;, "., .; Co m ments" "" ". :t .. Chuck Elmore Po s t, Buckley, Schuh& 904-878-5904 Je rnign, Inc . Steve Roskranse The R even u e Markets, 914-626-8655 General Manager Inc. Brian Taylor In t ernational Road 306-653-6611 D irect or of sales D ynamics, Inc. Richard Doering TransCore 619-552-4763 Steve Pustelynk OOCEA 407-425-8606 Orlando Oran ge County Expressway Authority Don Halvorsen AMP, Inc. 610-650-1580 Head of marketing Sally Oriffits TransCore, PA 717-561-5831 Oil McGlaughlin T ransCore, P A 717-561-5831 In charge of testing the T raffic 2000 treadle on the Pennsylvania Turnpik e Glyn Roberts Traffic 2000 0181 -948-6736 Invent or of Traffic 2000 piezo-electric Treadle and founder of Traffic 2000 Tony Lambert Traffic2000 0181-948-6736 Independent consultant hired by Traffi c 2000 Gorge Figuerado OOCEA 407-425-8606 Head of operations, Orlando Orange CoWlty Expressway Authority AI Palmer Palmer & Associates, 904-378-5904 Toll industry expert Inc. Bruce Anders o n Trans C or e Orlando, Fl Perform work for OOCEA DanAbshur AMTECH 214-733-6058 Te chnician working wi t h the Traffic 2000 treadle Tom Wagner TollTex 201-579-1776 Working on SWlPass Evaluation of T raffic 2000 Piezo -Eiectric Treadle 28