Application of real-time traffic adaptive signal control on the R44 arterial, stellenbosch

Mohr, Wilko Stefan (2019-12)

Thesis (MEng)--Stellenbosch University, 2019.

Thesis

ENGLISH ABSTRACT: The ever increasing level of traffic congestion experienced within and surrounding the town of Stellenbosch, South Africa, has been one of concern for many years. A large volume of commuter traffic creates a major influx of vehicles into and through the town during the morning peak traffic hour and, vice versa, a large outflow during the afternoon peak hour. These extensive traffic volumes exceed the provided capacity, thus creating an imbalance relating to the capacity vs demand ratio. This in turn leads to extended travel times, increased vehicle delay and a general inconvenience to the typical road user. Considering this and understanding that congestion has a negative effect on the individual road user as well as on the economy in general, this research project investigates the potential feasibility of using a Traffic-Adaptive Signal Control (TASC) system to optimise and manage the flow of vehicles through a series of signalised intersections. Such a system relies on real-time traffic data input and a representative traffic model which simulates and tests several signalisation options and, based on the objective function of the optimisation method, results in a more suitable allocation of green-time shares within the network. This, in essence, optimises the flow of traffic on individual intersection level while at the same time coordinating the traffic flow on a network level. The main objective of this research project, therefore, was to evaluate the applicability of a TASC system on the R44 arterial surrounding Stellenbosch in order to address the question of how applicable and effective TASC is in a developing world environment as a means to alleviate transport related problems on a traffic corridor. PTV Epics and Balance, proven to be successful in developed world application, is one such TASC system and was used within this study to address the research question. In order to realise useful results, a calibrated and validated simulation environment had to be created. Modelling was firstly done within PTV Visum to create an underlying macroscopic model of Stellenbosch in general, after which another PTV Visum subnetwork model of the R44-specific study area was generated. The final model, a PTV Vissim microscopic model, represented the evaluative environment within which detailed results could be obtained for a number of different scenarios. These scenarios where firstly aimed at identifying whether the TASC system would bring any improvements at all and secondly under which level of TASC control, inclusive of different vehicle detection options, the best results were obtained. Different levels of TASC control refer to exclusive local optimisation, exclusive global coordination or a combination of both. Detection options included a suboptimal layout of the existing inductive loop detectors as well as more consistent TrafiCam x-stream virtual loop Results comparison over a range of traffic flow parameters showed that the TASC system did indeed enhance the traffic flow. This was most notably found to be true under condition of the consistent camera detection and, predominantly, by using combined local and global TASC control. Based on the findings, this study recommends, with further study and continued expansion, the implementation of the TASC system on the R44 arterial, Stellenbosch.

AFRIKAANSE OPSOMMING: Die steeds toenemende vlak van verkeersopeenhopings wat binne die omgewing van Stellenbosch, Suid-Afrika ervaar word, was vir baie jare 'n bron van kommer. 'N Groot hoeveelheid pendelverkeer veroorsaak 'n groot toestroming van voertuie na en deur die stad gedurende die oggend spitsverkeeruur, en omgekeerd, 'n groot uitvloei gedurende die middagspitsuur. Hierdie aansienlike groot verkeersvolumes oorskry die voorsiende kapasiteit, en dit skep 'n wanbalans met betrekking tot die verhouding tussen kapasiteit en vraag. Dit lei dus weer tot verlengde reistye, verhoogde voertuigvertraging en 'n algemene ongemak vir die tipiese padgebruiker. As ons dit in ag neem en begryp dat opeenhoping 'n negatiewe uitwerking op die individuele padgebruiker sowel as op die ekonomie in die algemeen het, ondersoek hierdie navorsingsprojek die moontlike uitvoerbaarheid van die gebruik van 'n Traffic Adaptive Signal Control (TASC) -stelsel om die vloei van voertuie te optimaliseer en te bestuur deur 'n reeks gesignaliseerde kruisings. So 'n stelsel is afhanklik van intydse verkeersdata en 'n verteenwoordigende verkeersmodel wat verskeie signaliseringsopsies simuleer en toets. Op grond van die objektiewe funksie van die optimeringsmetode lei dit dan tot 'n meer geskikte toewysing van groentydse verdeling binne die netwerk. Dit optimaliseer in wese die verkeersvloei op individuele kruispuntvlak, terwyl dit terselfdertyd die verkeersvloei op 'n netwerkvlak koördineer. Die hoofdoel van hierdie navorsingsprojek was dus om die toepaslikheid van 'n TASC-stelsel op die R44-arterie rondom Stellenbosch te evalueer ten einde die vraag te bespreek hoe toepaslik en effektief TASC in 'n ontwikkelende wêreldomgewing is as 'n manier om vervoerverwante probleme te verlig. PTV Epics en Balance, wat suksesvol bewys is in die ontwikkelde wêreldtoepassing, is een van hierdie TASC-stelsels en is binne hierdie studie gebruik om die navorsingsvraag aan te spreek. Om nuttige resultate te realiseer, moes 'n gekalibreerde en gevalideerde simulasieomgewing geskep word. Eerstens is modellering binne PTV Visum gedoen om 'n onderliggende makroskopiese model van Stellenbosch in die algemeen te skep, waarna 'n ander PTV Visum-subnetwerkmodel van die R44-spesifieke studiegebied gegenereer is. Die finale model, 'n PTV Vissim-mikroskopiese model, verteenwoordig die evalueringsomgewing waarbinne gedetailleerde resultate vir 'n aantal verskillende scenario's verkry kon word. Hierdie scenario's was eerstens daarop gemik om te identifiseer of die TASC-stelsel enigsins verbeterings sou meebring en tweedens, onder watter vlak van TASC-beheer, met inbegrip van verskillende detektor opsies vir voertuie, die beste resultate behaal is. Verskillende vlakke van TASC-beheer verwys na eksklusiewe plaaslike optimalisering, eksklusiewe globale koördinasie of 'n kombinasie van beide. Detektor opsies het 'n suboptimale uitleg van die bestaande induktiewe lusverklikkers ingesluit, sowel as meer konsekwente TrafiCam x-stream virtuele lusverklikkers. Resultatevergelyking oor 'n reeks verkeersvloeiparameters het getoon dat die TASC-stelsel wel die verkeersvloei verbeter het. Dit blyk veral dat dit waar is onder die voorwaarde van die konstante opsporing van die TrafiCam x-stream kameras en veral deur gekombineerde plaaslike en globale TASC-beheer. Op grond van die bevindings, beveel hierdie studie, met verdere studie en voortgesette uitbreiding, die implementering van die TASC-stelsel op die R44-arterie, Stellenbosch aan.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/107305
This item appears in the following collections: