COMPUTER GRAPHICS
AND MICRO - COMPUTERS
povverful tools available to
transportation managers and planners
171,,MR Pi74'-,HRECEF3 77E'Di 7.7n/ZIUF.47E*-ni- F),sMi-2E CJIMN_T-P3 TIP7e,R_I-E- P'cULH'u [7;=?DREMR.iirM1 TF-7- fJ%ft_-77,57,'UtPrr 7[71qP,Z- -ceir FJ,F,EP-7,E17),RI'EU1E-F) 171-17/ 7[7,1ARE-Z7F3- -- PDERMCW__Egr,-=7'F) TIT-qe.,Wie)F,D177 FisRri-=2.,e-Frqrg, riTERE7E alfc -RF P-}uFzuLiwwirE- ar rLTF liD7ca Z-11U-1 - K[P,Fi -R-IMW,F_-,F=F7 Pprg„IW;----7c_nEUMF_=7,F, "ulg--)dr_leP-2)Uc- ,R,T7 riL -irEa 7F,r7Rui7=- ,=m77:4- 7,- DiTT-_!e `iiMe\F\ger-7MR1.- Dg3LF=HECi. ,==,,n)e,WZUPTZT[7:-17 7 t{[ wr7==4,p,.4_ceLLJIRI_IrT,‘,,fiEr7, Lir 77=u-„Ra zc--7_--r ÇRu 7ERRE-11:Eri7171 ("IjIkriEVF_- T-RenfiEFP [LfJE-m 7ReeriFFsZ.,R7i7 •E".--4- --R,7,E-17,-RF DipEca PFE,FELUWEIIE-Ell-R-A7MUZ-',FC757[1-1 rr 7act ----w\nraRzc q- 37, 7 e.\mceFuR-u- Vne f,[7HF-/%.Rf_. arr 5DÇL71F.1-7
,,k7: 11(-eF•CilP)7 7- r-CF4RH,3L -iirT :z,HZ3CL123[Wr4F_E- , 7F,1 i'RPR\il7a Fj.:r `11,717_!„WIMMM`ti- TEARrkla--irr `T'Fliempum -r=a IDirz- Fe) FrieiMEGfi "ERF-(77„-))GUE;FFe 7R/, ,7F,H).71 ce7F11- FL - ME- `17Rf-Uk,.1H,7]f - 7t,-' --- F2D (72 7 r\\_v US3P--(172L--,:7 F-1- 7 7E' a`c -tTME:' 71/ FFIUMFF,77FT-7, `7--r `11-Hafee- -,7R7R7r 177 1--LHJE. ,•c•777' UEZ, [7aF,-FciDucTuF ,:es. ee e \ -;E--7,12,E7FF HL Ller,,c:eVée:4ide, 17iii r-u-F rne":.",.,\E.S% ev- Aed'u'e&fgrf - 7JF,REci- A,,1\'-3FF2E-77 ‘4oe" e'4 de' 90 • . -u-Re.\[uerr7.1H1-1-. .-17[7 rtl Prt7,7-7, ee CWeeeee `i,--L7MHZFP=b7E,F., F-717=4 f=-7;Ftf"--Ti5C7•74nriLk,r, 43,eêe 7-7 r----JJEU P-DZL-MF=. 7E=;Rmr:17,e- -cr 17[7/:.\•,57-, 7f, ,-2F7 -177e.,R!1E,F•ZHJ7- C,à1r,
CANQ ' TR Gouvernement du Québec TTP Ministère des Transports 149A Direction générale du transport Ex.1 terrestre des personnes COMPUTER GRAPHICS 1 AND MICRO - COMPUTERS
powerful tools available ta transportation managers and planners
presentecl at the international transport congress MONTREAL • 1984
•-- -meer-ÉgÉbÈs:rnAlseoir-es3 ' Centre deldricurnentation:u 35, rLie,de+POrt:7R0yal'Est 3e étage Mdintréat.,110t4bçç): ,,Érry. 143Layl
Gouvernement du Québec Jean-Pierre- Primeau, ing . Ministère des Transports directeur de projets Direction générale du transport terrestre des personnes juillet 1984 PROJECT DIRECTOR Jean-Pierre Primeau, Eng.
COORDINATION Pierre Tremblay, Eng.
TEXT Jean-Pierre Primeau, Eng. Martin Nathanson, Eng. Pierre Tremblay, Eng.
TRANSLATION Martin Nathanson, Eng.
COLLABORATION André Babin, C.R.T. Claude Desloges Sandra Lavoie Robert Martin, Eng. François Mongrain Hien Nguyen Duc
GRAPHICS Denis Chauvette
TYPING NiCole Audet Louise Boivin SUMMARY SUMMARY
The purpose of this presentation is to describe the computer environment of which the "Direction générale du transport terrestre des personnes" has availed itself, and, as a result, to show how its management and planning activities are defined and carried out in this context.
The environment is characterized on the one hand by the use of micro-computers which, in addition to their local intelligence, allow access to a central mainframe'. On the other hand, computer graphics techniques play an important raie, which allows for the efficient representation of the enormous amount of data involved in transport management and planning.
These tasks can be classified under two major headings: data analysis, where the principal tools are the SAS and SPSS packages run on the mainframe whereas dBASEII, LOTUS 1-2-3 and Stat-Pac are used with the micro-computers; and transport modelling, aided by the EMME/2, MADITUC and UTPS software.
Both in terms of hardware and software, graphics and micro-computers have been effectively incorporated into our environment, facilitating considerably the work of planners and managers and enabling that work to reach a level of quality which could not be achieved using conventional methods. TABLE OF CONTENTS 11.
TABLE OF CONTENTS PAGE SUMMARY TABLE OF CONTENTS ii LIST OF FIGURES
INTRODUCTION 1
COMPUTER ENVIRONNENT 3
2.1 Hardware configuration 3 2.2 Mainframe 5 2.3 Local computers 7
2.3.1 "5280" system 7 2.3.2 PC-XT micro-computers 7 2.3.3 Tektronix 4116-13 8
2.4 Graphics devices 9
2.4.1 Tektronix 4116-8 9 2.4.2 Digitizer 9 2.4.3 HP/7475-A plotter 9 2.4.4 Calcomp plotters 10
PLANNING ACTIVITIES 11
3.1 Zone system 11
3.1.1 Subdivision 11
111
3.1.2 Digitization 13 3.1.3 Aggregation procedures 13 3.1.4 Node location 13 3.1.5 Zone equivalence 13
3.2 The data base 17
3.2.1 Origin-destination surveys 17 3.2.2 Socio-demographic data 17 3.2.3 Poils 18 3.2.4 Analytial tools 18
3.3 Auxiliary data 20
3-.3.1 Road traffic counts 20 3.5.2i Transit station counts 20 3.3.3 On-board counts 20 3.3.4 Itinerant vehicules 23 3.3.5 Travel times 23 3.3.6 Transit financial data 23 3.3.7 Car occupancy rates 23
3.4 Transport demand 25
3.4.1 Demand matrices 25 3.4.2 Demand projections 28 3.4.3 Modal split 28
3.5 Transport supply 30
3.5.1 Network definition 30 3.5.2 Network digitization 30 3.5.3 Transit une coding 31 3.5.4 Network analysis 33
iv
3.6 Simulations 35
3.6.1 Road network flows 35 3.6.2 Road travel times 35 3.6.3 Transit flows 35 3.6.4 Transit travel times 38 3.6.5 Iso-fare unes 48 3.6.6 Disaggregated itinerary analysis 41
3.7 Calibration 42
3.7.1 Road netword flows 42 3.7.2 Road travel times 42 3.7.3 Transit flows 43 3.7.4 Transit travel times 43
3.8 Other activities 45
3.8.1 Graphic production 45 3.8.2 SYRIC: Computerized system for carpooling 45 3.8.3 School transport study 45
4. CONCLUSION 47
ANNEXE 1: GLOSSARY 48 LIST OF FIGURES LIST OF FIGURES
PAGE
1 - Computer environment 4 2 - Hardware configuration 6 3 - General outline of activities 12 4 - Digitization of zone polygons 14 5 - Zone aggregations 15 6 - Survey data retrieval 19 7 - Representation of demographic data 19 8 - Station transit counts 21 9 - On-board counts 22 10 - Observed travel times 24 11 - Transit financial data 24 12 - Frequency distribution of generations totals 26 13 - Trip productions and attractions 26 14 - Trip demand corridors 27 15 - Evolution of transportation demand 27 16 - Digitizing environment 32 17 - Network coding 32 18 - Transit unes 34 19 - Transit capacity 34 20 - Traffic flows 36 21 - Frequency distribution of road travel times 36 22 - Road network isochrone unes 37 23 - Transit flows 37 24 - Metro ridership profiles 39 25 - Transit boardings 39 26 - Transit isochrone unes 40 27 - Analysis of fare structures 40 28 - Differences between simulated and observed traffic flows 44 29 - Ridership comparison at metro stations 44 30 - Carpool matching program 46 I INTRODUCTION 1
INTRODUCTION
In the contemporary context of urban public transport, where citizen participation has become increasingly important, where requests from the public and from their elected officials have become more and more demanding and where the costs are inexorably on the increase, the decision-making process has become more and more difficult.
It is therefore necessary that accurate contemporaneous information be available as a basis for assessing the current situation. It is also essential to possess tools which enable the impact of administrative decisions to be evaluated.
Since the effectiveness and the productivity of a public agency are directly related to the speed and flexibility with which its employees function, requiring them to work with unwieldy tools traditionally reserved for use by highly specialized professionals would serve no purpose. Moreover, the staggering quantity of resultats which these tools can produce can never be fully exploited nor efficiently be transposed into graphical form by manual methods.
In order ta maintain a high level of efficiency among its personnel and ta provide the best possible service to the transport community, the "Direction générale du transport terrestre des personnes" has followed a course of acquiring flexible tools incorporating graphics techniques and to render them accessible to ah l of its personnel.
It has therefore established an Information Systems Service to carry out the following tasks:
Acquire, maintain and analyse ail relevant data with respect to personal ground transport.
Develop or install models or procedures allowing for evaluation of road and public transport policy or projects 2
3) Design and maintain management information systems.
The presentation which follows will attempt to show the approach adopted by the D.G.G.T.P. in order to achieve these goals, with a particular emphasis on the role of micro-computing hardware and computer graphics techniques.
The first part of the presentation is a summary description of the computer environment which has evolved in our organization. This is followed by a demonstration, using numerous examples of graphics results, of how our activities are carried out in this environment. The reader will find, at the end of this text, a brief glossary of the acronyms used and reference index for the hardware and software products described. 2 COMPUTER ENVIRONMENT COMPUTER ENVIRONMENT
Our computer environment, which we have attempted to illustrate schematically in Figure 1, consists of an array of hardware facilities supported and driven by a variety of software. The hardware has been divided into three (3) categories: mainframe, mini and micro-computers graphics equipment.
The software cla ssification, considerably more complex, has the following general categories': operating systems communication systems high-level languages application software.
The illustration in Figure 1 is paradigmatic and while we recognize that the classification is debatable, our primary purpose is a descriptive one which will enable us to relate our actual transport activities to the computer tools used.
2.1 Hardware Configuration
Figure 2 is a schematic representation of the hardware configuration, where the pivotal point remains the mainframe operated by the Quebec Automobile Insurance Board (RAAQ) in Quebec City.
An initial telecommunications une links the mainframe ta our IBM 5280 mini-computer under the VTAM protocol. The 5280, in addition ta driving a une printer, allows for the emulation of IBM 3270 terminals. A number of PC-XT's can also carry out 3270 emulation under VTAM communications with the mainframe. 4 figure 1
Computer environment
. MAINFRAME
teDe'L 580/58 SAS e0 SPSS G PSS
IBm u-riLires
abieee
etk.•ka.a,/cee, ..0101t 04M
eece
,,\Is 3.8H twe, 150 / I SPF
*eeeeeee 4 cn 0 a • • 0 0 C 11 2. r M 0 0 0 33 1:7 4-e e /77 ri7 II .2 2 e
oct. ted01-
edo />, Noveda ave- Isile eekb.
/44 es», FI xiteriDc31- 5
A second telecommunications une links the mainframe to an intelligent Tektronix 41168 terminal under the TCAM protocol. Numerous additional communications links can be established locally using modems, which also allow our micro-computers accessibility to other computers and external data banks under the ASCII communications protocol (ex.: Ecole Polytechnique, CRT, employee personal computers).
2.2 Mainframe
As we have seen, the mainframe is an AMDAHL 580/5860 operating under MVS. We access this machine essentially using TSO and ISPF, although other methods are available. Similarly, although this installation supports several high-level languages, we use FORTRAN almost exclusively.
Of the principal application packages available on the AMDAHL, the most relevant to our purposes are:
UTPS - ("Urban Transportation Planning System"), developped by UMTA of the U.S. Department of Transportation. This is one of the most widely used transport planning packages in North America.
EMME/2 - (Equilibre Multimodal - Multimodal Equilibrium) is a relatively recent package, developped by the Centre de Recherches sur les Transports of the Université de Montréal and installed this year on our mainframe. It is an interactive menu-driven package designed to support ail network analysis and transport modelling functions.
MADITUC - (Modèle d'Analyse Désagrégée des Itinéraires en Transport Urbain Collectif). This package was developped at the Ecole Polytechnique de Montréal ta support functions related specifically to operational planning for transit systems. It is characterized primarily by its disagregated approach and by its data structures which provide for more effective use of information available from 0-D surveys. 6
figure 2
hardware configuration
IBM 5225 line printer
AMDAHL 580
mainframe
printer
L VTAM calcomp plotter L printer
printer TEKTRONIX
ç•I•1■ 11•••■• ••■ •■• ■■•. • • 4116 8 •
printer calcomp calcomp digitizer plotter
HP-7475A plotter cassette hardcopy
hard-disk 7
SAS (Statistical Anàlysis System) and SPSS (Statistical Package for the Social Services) are widely used statistical analysis and modelling packages.
GPSS - (General Purpose Simulation System) is used primarily for modelling stochastic phenomena, particularly traffic queues approaching highway interchanges and toll booths.
2.3 Local Computers
2.3.1 "5280" system
At our Montreal location, we operate an IBM 5280 "distributed data system", a mini-computer which, in addition to its computing function, allows the terminals to be emulated as IBM 3270's in communication with the mainframe and also drives a local line printer. This system includes an RPG compiler, enabling us to develop, for example, interactive programmes for survey data acquisition and validation. A complete budgetary and cost control system for the D.G.T.T.P. has been established in this way, as well as a data bank on Quebec Municipalities incorporating pertinent informations for the D.G.T.T.P. (ex.: city hall address, demographic patterns, licensed public transport operators, etc.).
2.3.2 PC-XT micro-computers
For some time, the D.G.T.T.P. has recognized the potential of micro-computers in transport planning and has moved towards making this technology available to its professional staff along with the proper software facilities.
Since their introduction, micro-computers have quickly become an integral part of the working environment and an indispensable tool in increasingly greater demand by our personnel. Our choice of machines were the IBM PC-XT's, each with 10 Mb of hard disk storage, and capable of emulating IBM 3270 terminals in communication with the mainframe. 8
Using modems a't 1200 baud, local communications can be effected with software such as PC-Talk or PC-PLOT. The latter package includes the very advantageous facility of allowing an IBM PC to emulate a TEKTRONIX terminal driven by the PLOT-10 language. The PC-XT's can therefore function as development tools for graphics applications which alleviates the heavy demand placed on the TEKTRONIX terminal for production purposes. As for languages, application programming is carried out principally in PASCAL, supplemented by the ubiquitous BASIC for the non-computer personnel.
Of the numerous application packages for the PC-XT available on the market, the following three (3) are primarily used:
dBASEII: data base management language;
LOTUS 1-2-3: speadsheet with graphics;
Stat-Pac: statistical modelling and analysis package.
It is noteworthy that some of the D.G.T.T.P. employees own micro-computers (es: Epson, Apple) and that these machines have been effectively integrated into our environment.
2.3.3 TEKTRONIX 4116-B
This graphics device incorporates the Intel 8086 (+8087) micro-processor, with 576 Kb of dynamic RAM. In this sense, and particularly since it has been equipped with a 10 Mb hard disk drive, it functions as a powerful micro-computer, under the CP/M-86 operating system executing programs developped locally. The only high-level language used for this development, apart from the IGL graphics subroutine library, is FORTRAN-86. The principal applications in local mode are the digitizing programs to which we will refer later. 9
2.4 Graphics Devices
2.4.1 TEKTRONIX 4116-8
This machine constitutes the pivotai point of our graphics applications. The screen is 630 mm (25 in.) with a visible pixel array of 4096 X 3120. Three (3) I/O ports provide R5232C interfaces to peripheral devices.
This apparatus can operate using TSO under TCAM communications with the mainfame or under CP/M-86 in local mode. Regardless of their origin, the graphics commands must conform to the PLOT-10 language in order to be interpreted by the graphic processor. The primary graphics applications are EMME/2 for transport modelling and SAS-Graph and SPSS-X for data analysis. There are two (2) graphics subroutine libraries invoked by our in-house applications. IGL, from TEKTRONIX, is general package which supports a wide range of computerized draughting requirements whereas QPLOT was developped at the Centre de Recherches sur les Transports of the Université de Montréal specifically for graphic processing of transport networks.
2.4.2 Digitizer
Another part of our equipment is an electromagnetic CALCOMP digitizing table with dimensions of 1100 X 1500 mm (44 X 60 in). This device is normally linked to the TEKTRONIX 4116-8 to carry out the digitization of networks and geographic zone boundaries, but numerous other applications are possible particularly since it can be driven by the IBM PC-XT's.
2.4.3 HP-7475 plotter
This small plotter, with maximum dimensions of 280 X 430 mm and six (6) addressable pens, provides for high resolution colour drawings. The graphics commands driving this device must 1 0
conform to the "HP-GL" protocol. In addition to in-house applications, in PASCAL or BASIC, the plotter can also produce drawings generated by LOTUS 1-2-3 from the PC-XT or by the TEKTRONIX functioning as the host.
2.4.4 Calcomp plotters
We also have access to CALCOMP plotters capable of producing drawings with four (4) colora on a 760 mm continous roll of paper. These devices recognize commands in the CALCOMP protocol only.
Interfaces have therefore been developped which allow the plotters to be driven by applications such as UTPS, EMME/2 and SAS-Graph or in-house programmes using QPLOT. The "Calcomp-Preview" package is also available, which emulates a Calcomp plotter for prior display and verification of the image on a TEKTRONIX terminal. 3 PLANNING ACTIVITIES 11
PLANNING ACTIVITIES
In figure 3, we have tried ta schematically show the general organization of activities related ta our transport data bank, according to eight (8) principal categories, within each of which various subjects are identified. We will summarily describe these subjects with particular reference ta the specific computer aids, which we have outlined previously, applied in each case. The list is not exhaustive, and indeed the headings used are not immutable since in reality the differentiation between subjects is far more fluid than it appears.
Of equal signifance in characterizing the organization are the two streams into which the activities fall. The first is data analysis, in the broadest sense. Software packages such as SAS, SPSS, LOTUS 1-2-3, Stat-Pac or dBASEII are used in this context ta produce statistical or descriptive analyses of the data.
The other stream cavera "transport modelling", where the objective is ta model the supply of transport (the networks) and the demand for trips, and then ta simulate volumes in the networks.
3.1 Zone system
3.1.1 Subdivision
Many of the activities related to urban transport planning require a cartographic subdivision of the region into analysis zones. However it is often the case that the subdivisions in different data banks are mutually incompatible thereby prohibiting the comparative analysis of data from different sources. Moreover the zone systems are frequently too detailed ta allow for rapid analysis of macroscopic phenomena. mi as am an ami MM MI es Mt win IM Me MI MI
figure 3
general outline of activities
I ZONE SYSTEM DATA BASE AUXILIARY DATA OTHER ACTIVITIES Traffic counts Graphic production Origin-destination Station counts Surveys Syric Subdivision On-board counts School transport DATA ANAUMS •4( Socio-demographic Itinerant vehicles
Opinion polis Travel times Financial statements of public CALIBRATION Employment transport operators Digitization Traffic flows Land-use Car occupancy rates Road travel times Transit flows Transit travel times
Aggregation TRANSPORT DEMAND
Demand matrices
Demand projections SIMULATION
Location Modal split Traffic flows TRANSPORT MODELS Road travel times TRANSPORT SUPPLY Transit flows Netword definition Transit travel times Equivalence Network digitization Transit une coding Iso-fare unes Network analysis Transit itineraries 11■11 13
3.1.2 Digitization
Digitizing the subdivisions offers a way of overcoming these constraints since the resulting data can then be manipulated by computer. In the first instance, the zone boundaries can be used as a back-drop to spatially distributed results produced by software such as EMME/2 or SAS (figure 4) as well - as by in-house applications as will be demonstrated.
3.1.3 Aggregation procedures
The main advantage to the digitization lies in the use of aggregation tables to create images of new zone systems simultaneously with a corresponding aggregation of the data. In this way, for the Montreal region, a rapid transition can be made from a 1500 zone system to a 700 zone system, 149 municipalites or 66 districts (figure 5). The graphics aggregation procedure is tied to the data representation structure which is based on three (3) lists: the vertices of the polygons, the edges of the polygons and finally the assignment of the edges to specific zones.
3.1.4 Node location
Another capability arising from digitization is the automatic identification of a zone associated with a network node or link pinpointed by a graphics input technique. Graphics input on the screen also allows for interactive creation of zonal aggregation tables or extraction of survey data for pinpointed zones.
3.1.5 Zone equivalence
This same procedure has enabled us to "recode" Statistics Canada data available et the enumeration area level according ta the MUCTC's base system of 1 500 survey zones. By 14
figure 4
digitization of zone polygons
LIGNES DE DEMARCRTION •1 DECŒPAGE DES ZONES DE L'EKULETE ORIGINE-DESTINATION 1982
MINISTERE CES TRANSPORTS DU OLEOEC DATE: 21JUN84 DIRECTION CENERRLE J TRANSPORT TERRESTRE DES PERSONNES MEURE:19.23.21 91141112 DE NIEES DE LA RECIONOE moNneu. PAR: F.MONCRRIN
BASE NETWORK efre. - .., , .,. ...... , . ..•
: .... .
.,,,,, .....
„.... s ,. .. s ...... ' .... -.....
, ...... , -...... \ , . \.„..-.
...... ...;
... . ., ...... ,, ...... :- „
.... .. , . .. .... . ,.... , . • ..... , .....,-.-4-
..... . ... vINDOw...... „.....›....f...... : .: LLa 5707,58284 . • uR. 8249/58508 . .i ... : ENN2/2 pRouEcT, DOTTP -- VALIDATION DES RESEAUV 1982 8 DATE. 84 82 » .13CENRRIO 1888. TRANSPORT EN COMMUN (VITESSES COMMERCIALES) 19 8 2 MODULE. 2.19 USERAS», A8 15
. figure 5
zone aggregations
LIGNES DE DENIFIRCRT ION DEODUPFGE DER DISTRICTS DE L'ENOUETE ORIGINE-DESTINRTION 1982 DGTTP ■ 40 il GO
fie
1 1 e sa efrit ■II reg104 illediAtelp rd 144 alik ..
MINISTERE DES TRRNSPORTS DU OUEBEC DATE: 21JUNS4 DIRECTION GENERRLE DU TRANSPORT TERRESTRE DES PERSONNES HE1RE:16.32M »MUE DE DGMEES DE LA REGION DE IIONTRERL PAR: F.MONGRAIN
LIGNES DE DEMPRCRT ION DGTTP DECOUPFGE DES MLNICIPAUTES Ou PLAN DE 7PANSPORT crcum-lsee Y là IMMO r1111103»11 MIIIIHM.OINUI
% UMM. Mirt-M1.1.08,1Mmon
MAIIWVULE ele
Millk.
Y"--....' MM ) if
MOLY f fiPlie
IIIIIR-1.1.0 •• / iii7tin-courlfrr
met.w.11.••
MNI.M. IM ■Nilm.._... ..1•11111 0 *Aie& ÉV.. NINISTERE DES TRANSPORTS DU OUESEC DATE: CIJ1A4 DIRECTION OENERRLE OU TRANSPORT TERRESTRE 0E6 PERSONNES HEURE:16.01.00 ORME DE DONNEES DE LE REGION DE PIONTRF_AL PRA, FAIONGRAIN 16
identifying the zone in which the centroid of each enumeration ares is located, a satisfactory recoding of the initially two incompatible systems is achieved, albeit with some inevitable overlaps. In the saine fashion, it is possible to associate zone numbers with postal codes, which produces an aggregation table useful, for example, in processing automobile registration files without having to geocode each record. 17
3.2 The data base
Urban tranport planning is dependant on two major sources of information: - origin-destination surveys socio-demographic data
3.2.1 Origin-destination surveys
Origin-destination surveys provide the means to not only qualify the demand for trips between various analysis zones but also to categorize and qualify that demand. We use primarily the O-D surveys carried out by the MUCTC in 1970, 1974, 1978 and 1982, which cover the Montreal region with a semple varying between 5% and 10% of ail households. Figure 6 shows a graphic example of data extracted from a survey.
We have carried out additional O-D surveys in order to study specific phenomena. Data acquisition and validation programmes have been developped for surveys on board suburban trains or express lane buses, to compile ticket counts for zoned fares and to process small O-D survey data collected in outlying suburban areas.
3.2.2 Socio-demographic data
The other major source of information is socio-demographic data provided by Statistics Canada or by the Bureau de la Statistique du Québec. Recoding programmes have occasionnally been used for compatibility between the zone subdivisions. The data includes, for example, population stratified by age and sex, and may serve to forecast future demand within the context of trip generation models. This category also includes employment and land use data provided by the Ministère des Affaires Municipales as well as information of an economic nature. 18
Demographic data can often be represented by density, as in shown in figure 7 which illustrates the population density by urbanized hectare for the city of Laval.
3.2.3 Polis
These data are supplemented by various public opinion polis or surveys concerning, for example, the fare structure for public transport.
3.2.4 Analytical tools
Whether it is for transport modelling or simply to respond to requests emanating from various parties concerned with transport in the region, the analysis of these data represents a significant part of our activities. The SPSS and SAS packages are used on the mainframe, while tools such as LOTUS 1-2-3, dBASEII and Stat-Pac are available - for the micro-computer. The actuel data acquisition is normally carried out on the "5280" system, using the RPG language, or on a micro-computer with dBASEII. Data from external sources are usually provided on magnetic tape. Complete familiarity with the data and software at our disposai allows us to quickly and easily extract any information requested, according to the specific requirements of the user. 19
figure 6
survey data retrieval
NOMBRE D'AUTOS PAR LOGIS REPENTIGNY
FRECRIENCY
8808
6080 4899 111111111
2
IIN9AUTO
SOURCE: BANQUE DE DONNÉES ENQUÊTE O-D 1982
figure 7
representation of demographic data
DENSITÉ NETTE DE POPULATION VILLE DE LAVAL
DENSITÉ 0 A 0.0 P/144 28-29.0 4/144 leeele:4 38.39.0 P/MA =1 4e A 49.0 4/64 50 A 59.9 •4.14 U1S111n 68 A 69.0 P/MA MM. 70 4 71.0 P/MA Min el A 89.9 P/MA
SOURCE: RECHERCHES ET STATISTIQUES VILLE DE LAVAL 20 I 3.3 Auxiliary data
A wide range of other data used for specific purposes are incorporated in our system.
3.3.1 Road traffic counts
By virtue of our access to information from the Quebec Ministry of Transport and the Traffic Service of the City of Montreal, as well as of other municipalities, have created a traffic 1 count data bank for calibration of our models. The survey stations are located in a grid pattern which allows the margin of error of simulations ta be evaluated and provides a means of comparing production/attraction measures from the O-D survey 1 with entrances and exits at the cordon points. 3.3.2 Transit station counts
For public transport, entrance and exit counts at metro and 1 train stations are available, as well as specific additional surveys such as movement counts at bus termini and at park-and- ride lots. These counts lend themselves ta comparative analyses with the results of simulations based on O-D surveys (figure 8). 1 3.3.3 On-board counts Counts on-board vehicules (trains or buses) allow the ridership along lines ta be assessed. Using the LOTUS 1-2-3 package with programmes written in PASCAL, a micro-computer procedure based 1 on ( 1 ) has been written for on-board ridership data acquisition and analysis (figure 9).
( 1 ) Chapleau, R. et Baass, K-G.; Use of Supercalc ta compile and report 1 statistics in public transportation, Ecole Polytechnique, Montréal, January 1984 (Presented at the 1984 Congress of the Transport Research Board, in Washington). 21
figure 8
station transit counts
Itinéraires 0—D vs Comptages NITRO - Ligne 1 20
15
10
t t
15
)
-20 Ho Re Le Cals Yi P9 Jo Pr Ir PaBd SL PI MC Pi Cet Ch is 1g Fr a 111 :dg Mont j\-,N Iiineraires 1111 ConsPia9.8 22
figure 9
on-board counts
C) CIL - PLAll DE TIMISPORT 1984 - 03SIGMA 44288 - COMPTES-A-BORD
Ligne Io. CIL: 60 Code DGITP: 460 Mos ligne: Direction: de RB Cap: (versaB/deHB/Est/Ouest) Dates: Jour: Sem Rom abrege: C-6003 (Sem/Sa/Di) Ligne CTL Dir Jour
ARRET Die- Passagers a Bord 'tance Teins PERIODE No. losud lo. Intereectico 0tml fan./ Pointe Creuse Pointe Creuse 24 x Utilisateurs Seq.C6TYP CTL ou incaliaation AR iour PH soir Meures Pte / 24 h
1 4269 1 Ter. KB 0.00 125 367 480 141 1113 54.36 2 4166 2 Laur/Cartier 1.45 134 399 500 150 1183 53.59 3 3247 S'Aar/Tourangeau 2.15 178 459 482 157 1276 51.72 4 3239 8 laur/Concorde 2.85 210 525 490 170 1395 50.18 5 3230 11 Laur/Gianchetti 3.63 204 589 473 177 1438 47.08 6 3228 14 St Martin/ Leur 4.66 256 740 515 209 1720 44.83 7 32/2 16 St lutin! Vallierres 5.08 274 777 512 212 1775 44.28 8 3220 17 St lutin! Sun Valley 5.37 278 832 510 209 1829 43.08 9 3219 20 St Rutin/ Industriel 6.22 252 817 506 206 1781 42.56 10 3215 21 St Martin/ Idiome 6.40 251 793 472 188 1704 42.43 Il 3216 22 St Martin/ Teaaier 6.85 247 772 442 179 1640 42.01 12 3217 23 C D Laval 7.76 229 723 442 187 1581 42.44 13 3172 24 Corbusier/St Martin 8.23 232 744 439 206 1621 41.39 14 3174 28 Carrefour/Corbusier 9.24 217 708 404 184 1513 41.04 15 3161 32 Carrefour Laval 11.01 144 378 327 166 1015 46.40 16 3162 35 Dan Johneoe/Fece LCHS 11.98 98 458 331 167 1054 40.70 17 4133 36C02000 12.58 84 442 304 163 993 39.07 18 4121 42 St Martin/Rayfield 14.01 35 404 252 141 832 34.50 19 4124 43 Labelle/Souvenir 14.63 23 406 225 118 772 32.12 20 4716 47 Motre Dame/Labelle 15.75 6 153 96 43 298 34.23 21 4718 50 notre Dane/75e Ave 16.29 4 79 55 14 152 38.82 22 4719 51 notre Dame/Coomedy 16.51 4 65 35 9 113 34.51 23 4132 54 Centre Hosp Juif Con, 17.15 0 0 0 0 0 0.00
3883 /Jour Eabarquents totaux: 428 1752 1176 527 64/Icor nombre de voyages: Il 25 14 14
60.67 Ombre de pais eoy/voysge: 38.91 70.08 94.00 37.64
heure:ale de charge: Periode CREUSE LE JOUR 09:018 à 15:300
Kou d'arret Io C.in : 1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 0 1 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 5 0 Tern MB 4269 367 IhIu110t110utIIIlhIhIIu IuIuuIluuuuhII Laur/Cartier 4166 399 •1111•4.1.1~1114•111141«111.“M•414~1** LiUrnOtkrajnen 3242 459 eneeememeeemeenteeemeemeemerottememeese Laur/Concorde 3239 525 iffiffM.Mf■t44.111.114.4.N1Wel.....».. Laur/Gianchetti 3230 584: mo*Heeme• St Martin/ Laur 3228 740 : meemeelifflememeemmeeeireeeemeteeemeeefeemmeeeremememeetessrememe St Martin/ Vallierr 3222 777 : erememememe*** St Bertin/ Sun Vaii 3220 832 mesemeeemememeameseeeseeemeemeemeeeemeemeremeffleemeemeemeteeneemeee St Martin/ Industri 3219 817 tttttt mammumenereweinimmearm***weefflifflefflumemmemmeeerafteeme St Martin/ Menama 3215 793 : femme». ******* »mem ***** eammetsiimmemememmes e St Martin/ Tessier 3216 772 . 1111411M•MMINM*MMINMINIMNINNIIMNINH. C D Laval 3217 723 : Ifflef11•111MM5500404I4INIHM00MIMMII1N1IMMIFIfflell* Corbusier/St Martin 3172 744 . f415400040001101.1144e.flffleeffIMINONNIM4M1111HIMMINFMMIN Carrefour/Corbusier 3174 708 . 1.• ■■ M*1111 flele500111111111111.1.11NIMMINWIIIMONNIM Carrefour Laval 3161 378 . 11111111011** Dan Johnson/Face LCD 3162 458 . C D 2000 4133 442 : *HUM" ***** 111411 1111••• 5t Martin/Mayfield 4121 404: Labelle/Souvenir 4124 406 : •amemmeemememeememeeeeeseememeeeteme. Hotte Daae/Lahelle 4716 153 : eemeeeemeemeee. Notre Dame/75e Ave 4718 79 : mmesseu Notre Dane/Chosedy 4719 65 : moule* Centre Hosp Juif Con 4132 0: lote: FACTEUR . 10.0
Doebre de voyages par periode (V) : 25 VOY/PER KM productifs (VEHIM.VeLZIGUEURLIGKE) : 428.75 VEM.KR Passagers embarques totaux par periode (E) : 1752 PASSAGERS Utilisation CUT.SUNCPASSABORD*DELTADIST” :9058.26 P0SS.18 Distance merise par passager (DX.UT/E) : 5.17 KM Indice de productivite kiloaetrique : 42.25 X CIPK.(UTM08101■C0PII•100) Passagers / KM productif (PAVKR.E/V08IM) : 4.09 PASS/KR 23
3.3.4 Itinerant vehicles
In order to develop volume/speed relationships, essential for estimating road traffic speeds at the peak period, a series of routes were travelled by itinerant vehicles operated by the M.T.Q. The resulting data, after processing, have led to a categorization of the major elements of the road network in the region and ta the establishment of a family of volume/speed curves which will be used for simulation purposes.
3.3.5 Travel times
Itinerant vehicles provide measures only of travel times along particular segments of the road network. However, in order to obtain a more complete verification of simulated inter-zonal travel times, a survey was carried out among two large employers in the region (Lavalin and the Université de Montréal). The processed data from this source can be represented by maps of isochrone unes showing the spatial distribution of road and transit network travel times (figure 10).
3.3.6 Transit financial data
Data related to the public transport fare structure have been gathered in order to build zone-ta-zone fare matrices. In conjunction with this, data has been collected according to a format that will facilitate analysis, using LOTUS 1-2-3, of ridership, revenue and operating costs of public transport operators (figure 11).
3.3.7 Car occupancy rates
Finally, surveys of private vehicle occupancy carried out at various observation points, are available for the morning peak period.
24 figure 10
observed travel times
LIGNES ISOCHRONES -- RESERU ROUTIER 82 DGTTP TEMPS OBSERVES SELON L'ENTE -- DESTINRTION L'UNIVERSITE DE MONTRERL
I
I 7
_ r.„,, fi ,,...,j11-/•-•-_,---7-4 `1 v, ‘-,-,, / ( , --- ' , ------ .1 \ / L...- / r (\_,) ', . «,•,,," / , ■
„, •.,--- ‘0 - ._ , 1
,, • r
.. ° ------ ,r ‘-" e ,,
nimisme DES TRANSPORTS OU OUEBEC ORTF: 23RU084 DIRECTION GENERRLE OU TRANSPORT TERRESTRE DES PERSONNES MEURE16.16.19 DANDUE DE DONNEES DE LA RECION DE MONTRERL PAR: F.MONGRAIN
figure 11
transit financial data
AUTOFINANCEMENT DES RESEAUX • Svolution durant ta pariade 1978 - 1984 70%
65% -
35% -
30% 1976 19'77 19178 1979 1980 1981 1982 1983 1984 1111NBES D C. T. L. + C.2C.U.M. o C. T.R. 25
3.4 Transport demand
Modelling on the basis of demand data assumes several forms.
3.4.1. Demand matrices
In the first instance, there are Origin-Destination trip matrices to be loaded onto the networks in order ta produce assignments. These matrices can be constructed for any system of zones defined by the user.
Graphical representation of these data is essentiel considering the enormous quantity of information which they reflect, and numerous methods are available.
Histograms, for example, can be produced to demonstrate the distribution of the number of observations in each element of the matrix. This provides a visuel indication of the number of elements which are empty or have negligible values and could therefore serve ta redefine the subdivision of the zones. A more effective technique of quickly evaluating the homogeneity of the zone system is a frequency distribution of system production and attraction totals (figure 12).
The trips production and attraction totals can be spatially represented by bars which immediately indicate the relative importance of trip generation (figure 13). The illustration of trip demand corridors becomes a simple, instantaneous task, allowing the planner ta effect numerous analyses with a minimum cost in terms of time and without constraints on the quantity of data which he can examine (figure 14).
A matrix can also correspond ta the difference tetween two other matrices. Hence the comparison of two demand matrices corresponding to different points in time will immediately reveal the zones of significant increase or decrease. Figure 15 shows the evolution of demand, in terms of trip production, for a part of the Montreal region between 1978 and 1982.
26 figure 12
frequency distribution of generations totals
PLOT H ISTOGRAM OF IIDE32: PROATT 5165175 5151 VILLES, errrn.
PREOUEMY
35 MOIM PROATT
32.5 OUT OP RAME BELLA8, 4515E, I: 38 WEIINT — ME 27.5 MEA/11136.7895 STO :1386.1948 25
22.5 _ 28 17.5 _ _ _1. ,2.5 _. I. .._ _r- 7.5 ■ Lfl
-.«. Lr
0 2511 588 TM 10010 1250 ISUS 1750 MM MOMs PMATT
EMME/2 PROJECTsDOTTA -- VALIDATION DES RESEAUX 1932 DATE, 84 86 21 MATRIM 8002, PROATT 501111E PRODUCTION ET ATTRACTION NMULE. 3.I6 USER:11630/ CD
figure 13
trip productions and attractions
PLOT MATR I X MF05: TCTBP 1 TOTAL 751 ORIGIN AND DESTINATION
'■!!0.' .50". Mr.• MWERU' m'yin
SCALEs 58
fir UMM. . .. . • 103118
8888
6055 I 4811 2888
-211110 • 85NDOIM 50513/583134 UR. 6108/50403
EMME/2 PROJECTsPLAN OE TRANSPORT EN COMMUN - VILLE DE LAVAL - DATE, 04 88 21 51E804I0 Os ANALVSE DE LA DEMANDE SUR I D IST RI C T S MOULEI 3.13 RAMIS *1055* TCTSPI TRANSPORT COMMUN TOUS BUTS POIN'E 1)555,8836/ CD 27 figure 14
trip demand corridors
PLOT MATR I X MPOS: TCTBR1 MATRI X BE on PAIRS ere2
SCALEI SA
4a
au la»
VIMY: IL: SESII/S8304 UR, 6188/68688 EMME/2 PROJECT, PLAN DE TRANSPORT EN COMMUN — VILLE DE LAVAL — DATE: 134 86 21 SCENARIO EN AAAAA SE DE LA DENANDE SUR 6 DIST. ICrS MODULE, 3.13 MATRIX AF1161 TCTEIP1 TRANSPORT COMMUN TOUS BUTS POINTE A... _USER:86313, CD
1 figure 15 evolution of transportation demand
PLOT MATR I X MF03: MT7882 TOTAL FOR SAISIR AND DESTINATION ere
SCALE: 180
18061 El 1311111 66111 H, 4431111
e
VINDOW: LL, 5EI28/61292 UR, 6261/68S42
E15IE12 PROJECT: COPPARAISON ENCUETE 00 1076 ET 1082 DATE, 64 16 22 SCENARIO la DECOUPROE EN 66 DISTRICTS SELON L . ENDUETE 00-113713 MODULE: 3.13 NATRIX AF831 EIT78112 DIFFERENCE 11382-107814PM USER.1630/ CD 28
3.4.2 Demand projections
Certain studies require the projection of demand for future time horizons. This procedure generally incorporates four (4) steps:
population projections employment projections trips generation trip distribution
Failing alternative means of forecasting, the previously described data base is used to develop a cohort survival model in order ta predict future population levels. For employment forecasts, data is provided by sources such as the OPDQ and MAM.
The traditional methods of aggregate trip generation have been rejected, since statistical analyses have shown that the socio-demographic variables available in the data are not sufficiently explanatory. As an interim alternative, linear projections of population cohorts and employment levels by category are used, and a disaggregate trip generation model is concurrently being tested.
A gravity model has been formulated for the purpose of re-distributing the trip matrix compiled directly from the O-D survey. Calibration of this model is currently in process.
We are also working with stochastic methods for temporal projections of demand matrices as a function of "observed" matrices which are available at four (4) previous points in time, corresponding to the O-D surveys in 1970, 1974, 1978 and 1982. 29
3.4.3 Modal split
We are not unaware of the currently extant problems inherent in the standard aggregate bi-modal split model, as our experience has shown that its explanatory power is limited. We are now developping procedures for estimating mode choice using, among others, the MADITUC package, in a disaggregate context. 30
3.5 Transport Supply
3.5.1 Network definition
Computer graphics software, particularly when it is interactive, has become an indispensable tool for coding transport networks. A computerized representation of the road network has been created incorporating highways, arterials and major collectors for the entire region of Montreal. In the same context, ail of the significant public transport unes have been coded for the morning peak period.
Although we use three (3) distinct software packages for transport modelling (EMME/2, MADITUC and UTPS), we have constructed a unique network which each package can transform into its required data representation. The public transport networks are constructed in terms of this base network. With this integrated approach, not only is data processing minimized but the sofware packages can be interfaced through I/O procedures. For example, the flow on a link simulated by MADITUC can become an attribute of the same link in the base network used by EMME/2. The management of this universal data base is carred out on the mainframe, by FORTRAN programmes or by TSO commands.
3.5.2 Network digitization
In order to code the networks in a systematic and efficient manner, we have developped an interactive data acquisition system using a digitizing table. Figure 16 illustrates the configuration of the system.
The digitizing tablet with the mouse is used to pinpoint network nodes (street intersections or transit stops) or ta trace non-linear network links. The graphic coordinates, 31
within the space of the tablet or the graphic display area of the video screen, are converted to the coordinates defined by the user according to the scale and origin of the superimposed map.
The command menu at the top of the video display provides a set of commands corresponding to the different keys on the mouse. Each key is programmed to activate a initial node location and numbering or link distance calculation.
The dialogue area at the bottom of the video display is used to send instructions and simple prompts to the user. The required answers are Y(es)/N(o) responses or alphanumeric information corresponding to node and link attributes (e.g. node numbers, street names, volume/speed categories, etc.).
As the network information is obtained from the digitizer, it is simultaneously displayed in the graphics area of the video unit as a visual aid to the user. In this way, node and link files, as diplayed in figure 17 by EMME/2, can be quickly generated.
3.5.3 Transit une coding
As we have mentioned, the public transport lines are coded on the universal road network (figure 18). Since each of the three (3) simulation packages has its own particular features with respect to transit une representation, a unique, integrated coding would not have been warranted. However, the integrated nature of the node and link data facilitates the task on analysis because familiarity with only a single network is required. Moreover it allows, as we have indicated, a transfer of assignment results, when required, from one package to another. 32 figure 16
digitizing environment
menu hardcopy graphic display area
dialo
Zn/MM plotter
GRAPHIC TERMINAL
TEKTRONIX 4116B
MAINFRAME
mouse HOST COMPUTER DIGITIZER
figure 17
network coding
BASE NE TWORK
mous. A
tM, :t33: LtAK n'us OWT I I 1 TE 90
:11.• .311-1 e.
:t13
-•"331 I 411 4e. e.r.1. œ --,ni, :EIt Port tu, de / , kr III tim Mortreal i t».ta.
"
:rH 1 :3[7:
....,....„,..., tic
t i• • ---/ t,r '7 H 44q, ,WE, Là, 6123/58397 UR: 8137/59480 1'41 ENNE/2 PROJECt.D0714 -- VALIDATION DES RESEAUX IO2 DAtE: 84 98 22 SCENAAIO I: TRANSAORT EN AuT011091LE .VENICULES. 1092 140311-E, 2.13 VSER:8830/ AS 33
3.5.4 Network analysis
Certain analyses specific ta the networks can be carried out graphically. In terms of the road network, certain types of roads, for example, can be extracted from the data bank and their transport capacities displayed cartographically.
From a transit perspective, service levels in terms of seating capacity on the lines can be shown (figure 19). A further possible application is the extraction of bus lines according ta criteria such as municipalities, the number of zones serviced or theresholds on the numbers of passagers transported. 34
figure 18
transit unes
TRANSIT LINES
LIN( TYPES. I TO 98 MIMES. MITAI MOISIT LINEls AU LI/88
Port I Mortreall
WINDOVs LLs 9121/9e311 UR, 8118/58487
EMME/2 PROJECT: - - D.O.T.T.P. - - BANQUE DE MINUS DE LA AMON DE NONTRE DATE: e4 OS 22 SCÉNARIO 2888: TRANSPORT EN COMNum -- 12 8 2 MODULÉ, 2.23 USERse8181/ 18
figure 19
transit capacity
TRANSIT LINES ON BASE NETWORK SEATED CAPACITY PER HOUR iefe LINK TYPES, 1 TO 041 TRANSIT LIMES, ALL LINES
SEULE. 3811
IIPT.Ple"b plegmq imam' Peint
WINDOWs Lls 8825/91442 UN. 8893/511482 EMME/2 PROJECTs - - D.8.T.T.P. - - SAME DE DONNÉES DE LA REOION DE MONTREAL DATE, 84 Se 22 SCEMARIO 21108s TRANSPORT EN COMMUN -- 902 MODULE, 2.23 _USER:1610/ 35
3.6 Simulations
Transport simulation consists of the prediction of network flows resulting from the convergence of demand with supply.
3.6.1 Road traffic flows
Traffic simulations estimate equilibrium flows on the principal components of the road network. The flows can be graphically represented by bands of widths proportional to the volumes on the links chosen (figure 20).
3.6.2 Road travel times
Since travel times and volumes are related in term of the volume/speed functions, the analysis of each is inextricably linked with the other. The travel time frequency distribution can be studied either in the form of a histogram or as a cumulative curve. Figure 21 represents weighted frequencies for car trip demand.
For the analysis of travel times, one of the most effective graphics techniques has been the plotting of isochrone unes. Depending on whether a row or column is extracted from the inter-zonal time matrix, contours of equal time, respectively from any production zone or any attraction zone to ail other zones, can be created (figure 22). These images can also be represented in three-dimensional from, by defining the time values as "altitudes".
3.6.3 Transit flows
In the same manner, the flows on transit corridor can be mapped with a clear illustration of the ridership levels (figure 23). This illustration can also be restricted to specific ridership on particular unes.
36
figure 20
traffic flows
AUTO VOLUMES
L IMK TYPES* ri ree,f ,ff . TO .. if_■4ref# ,..,_ / 4 Port w. aà t, . mo,,ea , Lr■ /,,, .4/ . irire,fr-„; ,ir. .. w ie • sc? P, t ff,, :.. ,,i. . e_ ...... : ,:„,,,.... :, .Lse s SEAUX 892 8 u"'",s, 6.: ortie TOM09ILEoss cVEMICULES1 1982
figure 21
frequency distribution of road travel times
PLOT HISTOGRAM OF MF02: CARTIM DERSITT CACTUAL VALUES, efe2 414EOUENCV CONSTPAINT. ',FOU. [ARISA li lir LOVER, 3 1 119811 UPPER. 159 DCUJDE
1 lir 1. 8402. CARTIN OUT OP RANCIE u. .1" BELOWo A 8900E, 8 VETO« Aret, AUT892 1911 ME68838.822882 500 t 19.91158
5951
088
3991
2911 Mils.
18418
• 2e 40 SI 89 100 12e 140 1C121 CARTIM
EI9E/2 PROJECT. D OTTP -- V AL I DA T ION ass RESEAUX 9622 DATE, 84 08 21 MATADI *FUI CARTIN MATRICE DES TENPS AUTOS NODULE, 3.18 USERAIS», CO I 37 figure 22
road network isochrone unes
LIGNES ISOCHRONES -- RESEAU ROUTIER DGTTP -- (LAVAL) TEMPS SIMULES EN POINTE OU MATIN ORIGINE VIMONT ,
I '
\ ' JP
I\ ■J,,,,,'
4 1 n \ / /....! r \ P \ -- - ,
, ‘1 _ „' - - — , __-,' \ , ! -,7 -7
NINISTERE DES TRANSPORTS DU OUEBEC DATE, ASER,» DIRECTION GENERALE DU TRANSPORT TERRES ,RE DES PERSONNES MELPER 9.05.22 NARGUE DE DONNEES DE LA REGION DE MONTREAL PARI F.MONDRAIN
figure 23
transit flows
TRANSIT AND AUXILIARY TRANSIT VOLUMES
LIN T'IPES. I TC 00 , THRESNOLD. • • LOYER. 1008 UPmEN• Namur
WINDOW. LLt 6177111101 UMM 4210128207 ENNEY2 ANOJECit - — D.O.T.T.P. — ANNALE DE DONNEES DE LA 199109 0€ NOME& I DATE* 04 NU ACENARIO 2111, TRANSPORT EN COMMUN - — 1002 MODULE, 4.22 AI 38
Figure 24 shows the ridership profiles for the morning peak period on three metro Unes of the MUCTC. These volumes correspond to itineraries described in the 1982 O-D survey, as validated by MADITUC. The figure was produced by SAS-Graph using MADITUC files.
Another interesting illustration demonstrates passenger movements at network nodes. Examples are initial boarding volumes, transfers, total boarding volumes, alighting volumes, etc. These schematic images immediately reveal not only the zones where ridership is most important but also the major demand corridors, where proposals for infrastructures of medium or high capacity would be appropriate (figure 25).
3.6.4 Transit travel times
In contrast to the road network, travel times public transport are not dependant on flows in the infrastructure so that the simulation results are really a replay of the initial "Transport Supply". Nevertheless, the same analyses as for the road network apply here, ie:
travel time distribution isochrone unes (figure 26) 3-D time surfaces
3.6.5 Iso-fare unes
Our programme for plotting iso-fare lines incorporates a procedure to analyse the costs and levels of service of public transport. The procedure calculates and maps break-even fares, in terms of demand and the allocation of operating and capital costs to transit users according to various formulae. Figure 27 illustrates, for the CBD, the equilibrium fares for a scenario where the user would defray 75% of operating costs and 25% of capital costs with the deficit supported by government subsidy. 39 figure 24
metro ridership profiles
RÉSEAU MÉTRO CTCUM
Henri-Bourassa
" 11,11.1i Honoré-Beaugrand ( tele , Ligne 2 Ligne 1
Plamondon
Berri-De Montigny
Ligne 2 Longueuil
Lionel-Groulx Ligne 4
Angrignon Ligne 1
figure 25 transit boardings a.m. peak period
TRANSIT VOLUMES AT NODES TOTAL 80ARDINSS . .i : LINE TYPES: .. 1 TO 60 TNRESHOLD: UPPERs ....d. .. ...*
4f . i ,•■ • .• ...... ,• ... ••••.
• • d• ... ;.
m.•• • ''', .• SCALE: 3 .2. • ,. ', :: O. g • • '
. • ...... ,, ., •
.. . ' 1 •
,
.• • ...... • .. • .••• . ' : • ...... • . .. WINDOW: . LL: 6087/58441 ..... -•'...... .. el 6149/50554 L . .4' • EMME/2 PROJECT: - - D.O.T.T.M. - - BANQUE DE DONMEES DE LA REDION DE MONTREAL S ONTEI 84 88 86 SCENARIO lems TERNSPORT EN COMMUN - - - le 8 2 - - - MODULE: 6.22 USER:8538/ p ■ 40
figure 26
transit isochrone unes
LIGNES ISOCHRONES -- TRRNSPORT COLLECTIF TEMPS SIALLES EN POINTE OU MATIN POP UTPS -- DESTINATION CENTRE-VILLE DGTTP
. ; \
re -, à r g Il , ; I, Vreji
• /' \ \ . A 1 • f' / 2,\./..... / ,--,/ ;----; ./.. i ..... 1 • G › g7 ,...., Il
,r- ' r I ...------, 7 f g.; MINISTER( 0E9 TRANSPORTS DU OUEREC DATE , muges DIRECTION GENERALE OU TRRASFORT TERMSTRE DES nRscffles MEURE 9.03.32 • SAME DE DONNEES DE LA %MM CE MONTREAL • PAR: F MONGRA(N figure 27
analysis of fare structures
LIGNES I SO- TRR I FR IRES CDONNEES 1978) INTERVALLE $20. ICARTE-MENSUELLE/ -- DESTINPTION CENTFE -VILLE DGTTP -'11Z-
,
_
,.... 1,
I I f (/'
-, ......
-,y ---
I%
,..., ' ,•• •,...„s—;•-•
-,
■ . , • . ,--- J-..; ,,,...... ) DIRECTION GENERALE OU TRANSPORT TERRESTRE CES PERScPoiES CATE. 2S.02E4 CENT MUE-MENSUELLE . NO. MOYEN DE DEFLACEMENTS (44 / MOIS) • TARIF DE USE ME4WE:11.53.4e HYMNE« DE TARIF. 500 COUTO D'IMADOILISATICN - 750 COJTS D'EXPLOI/RtION PRR. M.N./F.M. 41
3.6.6 Disaggregated itinerary analysis
Although not by definition a simulation package, the MADITUC software, which has been recently installed, serves as a basis for disaggregate analysis of trip itineraries arising from the MUCTC 0-0 survey. Among other purposes, this facility will be applied to studies of choices between modes of access to metro and suburban train stations as well as to the conventional modal split between car and transit. 42
3.7 Calibration
The calibration of transport models is a distinct stage of the modelling process in which simulation parameters, including the volume/speed functions, are adjusted by comparison of simulated results with those observed in the networks during the period to which to the data base corresponds.
3.7.1 Road traffic flows
The differences between simulated and observed flows can be graphically mapped in order to rapidly assess the extent and spatial distribution of errors in estimation (figure 28). This is supplemented by report generation procedures which produce the results of successive calibrations with indices measuring the performance of each step.
3.7.2 Road travel times
Another means of identifying weaknesses in the model is by comparison of the simulated and observed times in the network, of which one representation may be histograms of the differences between both time matrices.
The comparison of isochrone plots for observed and simulated times can serve to identify those corridors where adjustments should be made. Finally, the spatial representation by zone of the discrepancies in travel time will clearly high-light those sectors where there are significant errors of estimation. 43
3.7.3 Transit flows
The availability of entry and exit counts at stations allows for the creation of histograms which are useful in evaluating the performance of the assignment models (figure 29).
Similarly, by superimposing simulated riderships profiles on the corresponding observed profiles, the difficulties in the modelling process are isolated. However, it should be noted that the temporal heterogeneity of the observed counts, the surveyed 0-D data and the simulation results is a mitigating factor which complicates - the calibration of the models.
3.7.4 Transit travel times
As in the case of the road network, the estimated transit travel times can be compared to observations in the following formats:
distribution of differences comparison of isochrone maps spatial representation of differences 44
figure 28
differences between simulated and observed traffic flows a.m. peak period
AUTO VOLUMES DIFFERENCE SCENARIO 2 - SCENARC ereF
LISE TYPES: 1 70 58
SCALES 208
I COO panama.% SX
plaint
01140011: ! LL: 6037/58335 ... f UR, 6107/5EPI55 E1!E,2 PROJECTIDOTTP -- VAL:DATION DES RESEAUx 082 DATE: 04 86 86 SCENARIO 21 COMPARAI SON DES COMPTAGES ISCENARIO 21 VERSUS LES VOLUNES SIMULE mODULEo 8.13 ICENARIO Io TRANSPORT EN AuTORO BILE ovERIEUUS) 188 2 25E2,8630/ AB
figure 29
ridership comparison at metro stations
Sortants, Métro Ligne 1 Période de pointe du matin 24 22 ; 20
02 18 02 0 % 02 le 0% 0% 0% al" 14 1 cru OÎP 1 c 0 0% a 12 — 02 P 0%. o .c 00\ It%1 it101 4Ê, 10 202 (02 it001 8 1:1 W it: W 0002 109 0 2 05e% 02 l b 11 4 2 110. 2 ilo 1 02 , to, e e 2 t'■111,411Pgritit„ o Il 1!111:111!!1111;!111!! ii! ! estwille !ma '0.. 1,c d!! it!. Ho Ra La Ca As VI P9 Jo Pr FrPaBdSLPAMGPI Gy Aw Ch Ls Eg Vr JI MkAg
Stations IZZ Comptes (‘-"Çj itin.ba g3SD UTPS 4 CONCLUSION 45
3.8 Other activities
There are several other activities which we carry out generally in response to specific requests, and which do not fall into any of the above-mentioned categories. The following three examples suffice to illustràte their variety:
3.8.1 Graphic production
We are occasionally required to generate specific illustrations to be included in various reports or presentations produced by the D.G.T.T.P. Sofware such as LOTUS 1-2-3 provides flexible tools for responding efficiently to these requirements.
3.8.2 SYRIC: Computerized carpooling system
SYRIC is an in-house application for carpooling developped -for "Covoiturage-Québec" and with their assistance. The package written entirely in dBASEII, is menu-driven on a micro-computer, with simple dialogue between the user and the screen (figure 30). The package is currently installed and available at "Covoiturage-Québec".
3.8.3 School transport study
School transport constitutes an additionnai concern of the D.G.T.T.P. and a working group is currently examining the state of the school bus fleet in the province, for which a data base was built by merging information from the R.A.A.Q registration files with data obtained from financial statements of fleet operators with membership in A.T.E.Q. This first step was done using S.A.S. on the main frame, but ail subsequent analyses can be carried out locally on a micro-computer using the Stat-Pac or LOTUS 1-2-3 packages. 46 figure 30 carpool matching program
SYRIC SYRIC
ESSAI S 1
MENU DU PROGRAMME DE SAISIE Current date is Tue 1-01-1980 Enter new date: 6-30-84 Commence un nouveau fichier de participants Current time is 0:00:37.51 Continue d'ajouter des participants au fichier Enter nec time: 11 Revient au menu principal SYSTEME DE REGROUPEMENT
The IBM Personal Computer DOS INFORMAT . I.SE Version 2.00 (C)Copyright IBM Corp 1981, 1982, 1983 DES COVOITUREURS C)cd‘dbase
C)SYRIC Votre Prenom --):JEAN Nom --):GILLES VOTRE CHOIX 11 a 3)7:3: OT DE PASSE --):SYRIC
SYRIC
MENU DU PROGRAMME DE REGROUPEMENT
REGROUPEMENT DE(S):
Derniers participants inscrits au fichie saisie. Certain participants .choisis. Tous les participants non-groupes Bonjour JEAN Tous les participants connus Revient au menu principal ton dernier acces fut le 27/06/84 NOTE: les choix 3 et 4 peuvent necessiter beaucoup de temps. Date d'aujourd'hui ? (JJ/MM/AA) :30/06/84:
VOTRE CHOIX (1 a 5)7:5:
SYRIC
MENU PRINCIPAL MENU DU PROGRAMME DE MISE-A-JOUR
I) Ajout . de nouveaux participants Modification au fichier maitre des participants. Regroupement des covoitureurs Rayage d'un participant dans le fichier-maitre. Mise-a-jour ou modifications aux fiches des participants .3) Retour au menu principal. Editeur de rapports Fin de session
votre choix ---):1:
VOTRE CHOIX (de 1 a 51:5:
SYRIC
MENU DE L'EDITEUR DE RAPPORTS
Totaux sur les participants du fichier-maitre. Liste et modification des itinerairem de fourgonette et taxi Sommaire des participants. Rapports de groupement. Retour au menu principal.
VOTRE CHOIX 11 a 5)7:5: 47
CONCLUSION
Three years ago, there were no more than five professionnals working on a regular basis within a computerized environment in our organization. But as the environment itself has evolved, particularly with micro-computer technology, it has absorbed a growing number of our personnel, bath professionals and techniciens, more than thirty of who, not including about ten secretaries capable of data entry tasks at the terminals, are now making extensive use of the tools available.
Many tasks are now effected interactively. Validation of data is done simultaneously with its acquisition and in many cases the analyses are carried out with interactive graphics.
We foresee extending this process of computerization to incorporate our library and our draughting facilites.
Computer graphics and micro-computers have become a part of our daily working environment. GLOSSARY 48
GLOSSARY' OF ACRONYMS USED IN THE TEXT
A.T.E.Q. Association des Transporteurs d'Écoliers du Québec CALCOMP California Computer Products Inc., Anaheim, CA 92801 CBD Central Business District CP/M-86 Digital Research, Pacific Grove, CA 93950 C.R.T. Centre de Recherche sur les Transports, Université de Montréal C.T.C.U.M. Commission de Transport de la Communauté Urbaine de Montréal dBASE II Ashton-Tate, Culver City, CA 90230 D.G.T.T.P. Direction générale du transport terrestre des personnes, M.T.Q. EMME/2 Équilibre Multimodal-Multimodal Equilibrium, from the C.R.T. G.P.S.S. General Purpose Simulation System, from IBM HP Hewlett-Packard Company, San Diego, CA 92127 HP-CL Hewlett-Packard / Graphics Language IBM PC-XT IBM Personal Computer, model XT IBM 5280 Distributed information system, IBM I.G.L. Interactive Graphics Library, Tektronix INTEL Intel Corporation, Santa Clara, California 95051 I.S.P.F. Interactive System Productivity Facility, IBM LOTUS 1-2-3 Lotus Development Corporation, Cambridge, MA 02142 MADITUC Modèle d'Analyse Désagrégée des Itinéraires en Transport Urbain Collectif, École Polytechnique de Montréal M.A.M. Ministère des Affaires Municipales du Québec M.T.Q. Ministère des Transports du Québec O-D Origin-Destination 0.P.D.Q. Office de Planification et de Développement du Québec PC-DOS Disk Operating System (IBM Personal Computer) PC-PLOT Tektronix 4010 Terminal Emulator, Microplot Systems Co., Columbus, Ohio 43229 PC-TALK Pc-Talk III Communication Program, Freeware, Tiburon, CA 94920 PLOT-10 Plot-10 Terminal Control System, from Tektronix 49
QPLOT Graphic subroutine library, from the CRT R.A.A.Q. Régie de l'Assurance Automobile du Québec RAM Randon Access Memory R.J.E. Remote Job Entry RPG Report Program Generator, IBM RS-232 Standard RS-232 (Electronic Industries Association) S.A.S. Statistical Analysis System; SAS Institute Inc., Cary, North Carolina 27511 SNA Systems Network Architecture S.P.S.S. Statistical Package for the Social Sciences, SPSS Inc., Chicago, Illinois 60611 STATPAC Statistical Analysis Package, David S. Walonick, 1983 SYRIC Système de Regroupement Informatisé des Covoitureurs, D.G.T.T.P. TCAM Telecommunications Access Method TEKTRONIX Tektronix Inc., Beaverton, Oregon 97077 T.S.O. Time Sharing Option U.M.T.A. Urban Mass Transportation Administration, Washington U.T.P.S. Urban Transportation Planning System, from U.M.T.A. VTAM Virtual Telecommunication Access Method