M.Com. (IT Management) / Vehicle traffic congestion on Gauteng’s metropolitan and national roads in South Africa remains a challenge, especially during peak hours and also when incidents occur. They hamper the flow of vehicle traffic. Traffic congestion has negative consequences for business and for commuters’ quality of life. The goal of this research project is to identify variables that influence the flow of traffic and to design a vehicle traffic prediction model which will predict the traffic flow pattern in advance, given a set of predictor variables that will enable motorists to make appropriate travel decisions ahead of time. The traffic flow was categorised into three classes, namely traffic jam, free flowing and flowing congestion. In this study, the artificial intelligence algorithms that include Bayesian Networks, K-Nearest-Neighbour, Multi-layer Perceptron (MLP) and C4.5 Decision Tree were used individually for predicting the vehicle traffic flow. The results obtained from these algorithms were compared using the predictive performance and prediction costs. The best predictive model is one that has lower cost and good performance. The results show that the MLP was the best performing algorithm in terms of predictive performance and low prediction costs. In order to predict a novel instance, a feed forward Multi-layer Perceptron network was built using Matlab and was used to predict the unseen vehicle traffic instance, also called a novel instance. The (MLP) model accurately predicts vehicle traffic flow on a single novel instance with a prediction performance of 80% (16 out of 20) on novel instances.

Thesis (MScEng)--University of Stellenbosch, 2008. / ENGLISH ABSTRACT: It is argued that an understanding of variability is central to the modelling of travel behaviour and the assessment of policy impacts and is not the peripheral issue that it has often been considered. There is a growing need to assess multi-day data to assess the distribution of user charges for road pricing, or patterns of public transport usage as well as improve the ability to identify mechanisms behind travel behaviour for modelling purposes. Drawing on studies worldwide, in conjunction with a review of the literature, the thesis first examines the relevance for using multi-day data, then discusses the methodology and results of a five week survey undertaken specifically for this study, makes a comparison of the findings with that observed in other studies, and finally discusses issues relating to the application of the data and future research possibilities. Previous studies have shown that behaviour which makes up the daily travel pattern can be highly repetitious in nature but that observing an individual’s behaviour on a single day might not be representative of their routine travel and that this behaviour varies across demographic segments and driver gender. This paper examines day-to-day travel behaviour variability of a residential area, Summer Greens, located in Cape Town (South Africa) using a travel dataset collected recently in November/December 2006. The survey technique employed was the recording of numberplates of all exiting vehicles from 06:00 to 10:00, weekdays from Monday to Friday over a period of five weeks. A total of 5677 vehicles undergoing 44 743 trips was observed and analysed. This research replicates and extends previous work dealing with day-to-day variability in trip-making behaviour that was conducted with data collected by Del Mistro and Behrens (2006) in Buitengracht Street, Cape Town CBD, in July/August 2005. The present research extends the earlier work by including departure time variations as well as conducting the observations during both a school period (3 weeks) and consecutive holiday period (2 weeks). Further, the thesis presents a method to measure day-to-day variability using the available data surveyed. This thesis finds a considerable difference in school and holiday traffic volumes as expected, but that despite this, certain identical travel behaviour patterns (such as vehicle appearance frequencies, following week repeat proportions etc.) is observed during both these periods. It was found that the peak hour for both school and holiday periods occurred during the same time period and greater traffic volume variability was found to occur on Fridays during the holiday period than in the school period. Traffic volumes across all weeks appeared to decrease from Monday to Wednesday and “bounce back” on Thursday and Friday consistent with the findings of another international study. It was found that motorists exhibited more departure time freedom during the holiday period with average holiday departure times much later than during the school period. Departure times were also observed to gradually become later from Monday to Friday during both the school and holiday periods with Thursday and Friday departure times significantly different to the other weekdays. The proportion of unique vehicles observed was found to increase with time of day and the resulting impact of this on the effectiveness of Variable Message Sign (VMS) applications is also discussed. The research concludes by applying the findings to determine the impact of a hypothetical congestion pricing scheme on traffic volumes. / AFRIKAANSE OPSOMMING: Die argument is dat ‘n begrip vir variasie sentraal staan tot die modellering van reisgedrag en die assessering van beleidsimiplikasies, en is dus nie net n nagedagte soos wat gereeld gedink word nie. Daar is ‘n toenemende behoefte om multi-dag data te analiseer om die verspreiding in gebruikerskostes te bestudeer vir die waardasie van paaie, of om verhoudings in publieke vervoergebruik uit te lig, asook, vir die verbetering van die vermoë om die meganismes agter reisgedrag te identifiseer vir modelleringsdoeleindes. Die tesis bestudeer eerstens die relevansie van die gebruik van multi-dag data in samewerking met die literatuurstudie, gebaseer op studies wêreldwyd onderneem. Tweedens bespreek dit die metodologieë en resultate van ‘n vyf-week lange opname wat spesifiek vir hierdie studie onderneem is en tref vergelykings met die resultate verkry deur vorige studies. Dit bespreek dan die probleme wat voortspruit uit die toepassing van die data en ook toekomstige moontlikhede tot verdere navorsing. Vorige studies het getoon dat die gedrag wat tydens daaglikse reispatrone voorkom van nature hoogs herhaaldelik is. Wanneer ‘n individu se gedrag observeer word op ‘n enkele dag is dit egter nie noodwendig verteenwoordigend van sy/haar roetine reisgedrag nie en dat hierdie gedrag afhanklik is van demografiese faktore en die geslag van die bestuurder. Hierdie tesis bespreek die variasie in dag-tot-dag reisgedrag van ‘n residensiële gebied, Summer Greens, in Kaapstad (RSA) deur gebruik te maak van ‘n datastel wat onlangs saamgestel is (November/Desember 2006). Die data is ingesamel deur ‘n opname te doen van die nommerplate van alle uitgaande voertuie tussen 06:00 en 10:00, weeksdae van Maandag tot Vrydag, oor ‘n periode van vyf weke. A totale aantal voertuie van 5677 wat 44 743 ritte onderneem het, is waargeneem en ‘n analise is uitgevoer. Die navorsing herhaal en brei uit op vorige werk wat die dag-tot-dag variasie in ritopwekking bestudeer. Del Mistro en Behrens (2006) het data bestudeer wat ingesamel is in Julie/Augustus 2005 in Buitengrachtstraat, in die Kaapse Middestad (RSA). Die huidige navorsing brei uit op hierdie werk deur die variasie in vertrektyd in te sluit asook om waarnemings te doen tydens die skoolperiode (3 weke) en die daaropvolgende vakansieperiode (2 weke). ‘n Metode word ook aangebied om die dag-tot-dag variasie te meet deur middel van die beskikbare data wat ingesamel is. Hierdie tesis vind ‘n merkwaardige verskil in die skool en vakansie verkeersvolumes, soos verwag kan word, maar ten spyte daarvan bestaan daar sekere identiese reisgedragspatrone (byvoorbeeld die frekwensies waarteen ‘n voertuig voorkom, weeklikse herhaling van proporsies, ens.) gedurende beide hierdie periodes. Die bevinding is dat die spitsuur vir skool- en vakansieperiodes gedurende dieselfde tydsperiode plaasvind en dat verkeersvolumes groter variasie toon op Vrydae gedurende die vakansieperiode, as die skoolperiode. Dit wil blyk of verkeersvolumes gedurende al die weke afneem van Maandag tot Woensdag en dan “terug spring” op Donderdae en Vrydae. Hierdie bevinding is in ooreenstemming met resultate van ‘n ander internasionale studie. Motoriste het ‘n neiging getoon tot groter vryheid in terme van vertrektye tydens die vakansieperiode, met die gemiddelde vakansie vertrektye heelwat later as gedurende die skoolperiode. Vertrektye is ook gevind om geleidelik later te word van Maandag tot Vrydag gedurende beide die skool en vakansieperiodes, met Donderdag en Vrydag se vertrektye wat merkwaardig verskil van ander weeksdae. Die verhouding uniek waargenome voertuie het volgens hierdie navorsing se bevindinge toegeneem met die tyd van die dag en die gevolglike impak hiervan op die effektiwiteit van sogenaamde “Vehicle Message Sign” toepassings word bespreek. Die navorsing sluit dan af deur die bevindinge toe te pas om die impak te bepaal wat ‘n hipotetiese kongestieprysskema op verkeersvolumes het.

M.Ing. / Link performance curwes represent the relationship between the travel time and the traffic flow on a link in a traffic network. This relationship is an essential input for trip assignment procedures. A number of different forms of link performance curves have been suggested. These curves are defined as mathematical functions (known as travel time functions) with link flow as the independent variable. Travel time functions usually have a number of parameters that may be varied according to link characteristics...

M.Phil. (Civil Engineering) / The SIDRA (Signalised Intersection Design and Research Aid) is calibrated for South African traffic conditions. Traditionally, South African traffic engineers made use of American or Australian based information to analyze traffic operations at signalised intersections. Two major models used in South Africa are HCM (Highway Capacity Manual) by the Transportation Research Board in the USA and SIDRA by the Australian Road Research Board. SIDRA is gaining popularity in South Africa because of the practical problems of the mirror image, which has to be maintained in using the HCM method due to the right side driving in the USA and other possible differences on driver behavior, vehicle attributes, geometric design and traffic control. SIDRA seems to be more suitable for South African conditions. Recently the use of SIDRA in this country has been increased to 44 organizations. This study aims to prepare a unified default file for South African users. Due to the large number of model parameters in SIDRA, priority is given to important ones for which information is readily available. And this study is confined to signalised intersections rather than all intersection types.

M.Ing. / Another method is suggested in this paper to decide on the method of control at intersections. A computer program has been developed to simulate traffic on a macroscopic basis to obtain cost for delay, fuel consumption, accidents, maintenance, oil consumption and capital investment. The total cost per vehicle proved that the existing criteria are conservative in so far they warrant traffic signs earlier than necessary ...

Thesis (MEng)--University of Stellenbosch, 2001. / ENGLISH ABSTRACT: Peak mornmg traffic-flow data were obtained from video footage of three representative freeway sections on the Nl and N2 westbound towards Cape Town. Flow, speed, and density measurements were made from the footage with the aid of a stopwatch. Many researchers (2-12) have originated and developed models to describe the relationships between traffic flow characteristics (speed, flow, and density) on freeways. In this report, a number of these models have been investigated with data obtained from South African freeways. The ability of each model to predict flow parameters over the entire range of data was evaluated with the aid of statistical methods. The tests were performed by regressing average speed vs. average density. Flow-density and speed-flow relationships were derived through application of the steady-state equation (2.6). In each case, a final model was chosen through visual inspection that consisted of two separate curves, one for the uncongested flow regime and one for the congested flow regime. Furthermore, speed-flow relationships were examined for individual lanes and compared to relationships established for average lanes. The models were also compared to models obtained from overseas studies (1,16,19) as well as from studies done locally (17). A secondary objective of this study is to investigate the performance of existing freeway facilities through application of the relevant models to the traffic-flow data of a particular facility. The current peak-morning performance of the N2 freeway section is investigated in terms of travel-time and travel cost. The particular study section consists of three lanes, the right hand lane being an HOY lane dedicated to taxis and buses. Different hypothetical cases are investigated. The first hypothetical case is an investigation into the traffic situation on the freeway section without the influence of the HOY lane. The second hypothetical case investigates the traffic situation on the section with perfect operation of the HOY lane. The current performance of the N2 section is compared to the performance of each of the hypothetical cases. / AFRIKAANSE OPSOMMING: Oggend-spits verkeersvloei data is verkry vanaf drie verteenwoordigende seksies op die Nl en N2 deurpaaie naby Kaapstad met die gebruik van 'n video kamera. Vloei, spoed, end digtheid opnames is gemaak met behulp van 'n stophorlosie. Verskeie navorsers (2-12) het modelle gepostuleer en ontwikkelom die verhoudings tussen verkeersvloei eienskappe (spoed, volume, en digtheid) op deurpaaie te beskryf. In hierdie verslag word 'n aantal van hierdie modelle ondersoek met data wat verkry is van Suid-Afrikaanse deurpaaie. Die vermoë van elke model om vloei eienskappe oor die hele bestek van die data te voorspel is geëvalueer met behulp van statistiese metodes. Statistiese toetse behels 'n regressie analise van gemiddelde spoed teenoor gemiddelde digtheid. Volume-digtheid en spoed-volume verwantskappe is direk afgelei vanaf Vergelyking 2.6. Vir elke geval is 'n finale model m.b.v. visuele inspeksie gekies wat bestaan het uit twee afsonderlike kurwes, een kurwe vir die vryvloei regime en 'n ander kurwe vir hoë-digtheid toestande. Verder word spoedvolume verwantskappe vir afsonderlike deurpad-lane ondersoek en vergelyk met verwantskappe wat verkry is vir gemiddelde lane. Die modelle word ook vergelyk met modelle wat verkry is vanaf oorsese studies (1,16,19), sowel as met modelle wat plaaslik verkry is (17). 'n Sekondêre doel van hierdie studie is om die prestasie van bestaande deurpadfasiliteite te ondersoek deur die verskillende modelle aan te wend tot die verkeersvloei data van 'n betrokke fasiliteit. Die prestasie van die N2 deurpad seksie gedurende oggend-spits verkeer is ondersoek in terme van reistyd en ryskoste. Die betrokke seksie bestaan uit drie lane, waarvan die regter laan gereserveer is vir busse en taxis. Verskeie hipoteses is ook ondersoek. Die eerste hipotese is 'n ondersoek na die verkeers-vloei kondisie op die seksie sonder die invloed van die bus- en taxi-laan. Die tweede hipotese ondersoek die seksie met perfekte werking van die bus- en taxilaan. Die huidige prestasie van die N2 seksie is vergelyk met die prestasie van elk van die hipoteses.

Submitted in fulfillment for the degree of Magister Technologiae: Engineering: Civil, Durban University of Technology, Durban, South Africa, 2013. / An increase in road traffic, poor road conditions and high numbers of road accidents are major challenges at the Port of Durban. Roads are considered as the most important transport mode at the Port of Durban. Road transport has taken almost 80% of the import and export cargo while railway transport is left with approximately 20%. It is estimated that 75 million lives in the world will be lost and 750 million people injured in road accidents in the first half of the 21th century. The Port of Durban is an important vehicle for facilitating economic growth of local, regional and national industries. For the Port to maintain global competitiveness with the current trend of globalization, it has to ensure that roads are well maintained, safe and have a smooth traffic flow with no delays. This study provides an overview of the road infrastructure within the Port of Durban in relation to road condition, safety, law enforcement and traffic. Existing and ongoing studies conducted in South Africa and abroad form part of the literature review. This study identifies factors that are affecting the condition of road infrastructure such as growth of container cargo, an increase in the dimension and weight of trucks, transport deregulation, port layout and handling equipments. Deregulation of road transport over the past years has resulted in an 80:20 split between road and rail transport putting more pressure on roads. Cost and time are the major deciding factors in the freight industry. Most customers prefer to use road transport due to the lower cost and reduced time compared to rail transport. There are eight major roads within the Port that connect the South, West and North of eThekwini Municipality namely Bayhead, Quayside, Maydon, Rick Turner, Wisely, South Coast, Bluff and Iran Roads. Asset verification and assessment of the condition of the existing eight major roads found that Quayside Road is in a better condition compared to the other roads. Maydon and South Coast Roads are low rated roads which are in a poor condition. Identification and assessment of the condition of 210 000m² of asphalt paved areas which included minor roads within the Port of Durban was also conducted. Most paved areas and roads fall under D (fair) category which is reasonable but maintenance work may be required within six months. Comparison between the condition of the eight major roads within the Port and outside the Port was investigated. The findings indicate that sections of roads outside the Port are in a better condition than sections within the Port. Traffic counts were conducted in order to determine the utilization of the existing eight major roads. Bayhead and South Coast Road are highly utilized roads. Road accident reports and death reports were analyzed on these roads. Most of the road accidents take place on South Coast Road. A questionnaire survey was conducted, targeting road users who travel on these roads within the Port. Feedback was obtained on the status of road conditions, safety and traffic within the Port of Durban. Findings of this survey revealed that most of the respondents don’t know where to report road defect/s within the Port. Approximately 37.5% of the road users felt not safe to drive on roads within the Port especially on South Coast Road. A high percentage of people (93%) witnessed accidents on these roads. Traffic signals within the Port are maintained by eThekwini Municipality and are very often non-functional. When road signs need to be repaired or replaced, it takes longer than expected. Also, there are limited parking areas around the Port resulting in trucks parking closer to the premises while waiting to collect or deliver cargo. This causes major traffic congestion, for example, on Maydon Road where most trucks park on the side of the road. Recommendations include assessment guidelines that could improve road condition, safety and traffic flow. Areas to be improved with regard to road infrastructure are also highlighted. / M

This thesis presents the results of an investigation into traffic operations and driver behaviour at traffic circles under South African conditions. The scarcity of local traffic circles necessitated the development of a simulation program (TRACSIM) to assist in the research process. This microscopic program for single lane circles is based on event updates and was calibrated and validated based on local data. Because the acceptance of gaps is such a vital part of the operation of a traffic circle, it was examined in detail. Specific attention was given to the possible use of a gap acceptance model based on variables other than time. Since the gap acceptance process also depends on the gap distribution in the circulating stream, the effect of the origin-destination pattern was also investigated. Two existing analysis techniques are evaluated and verified for local conditions, improving them where possible. Generally these techniques under-estimate traffic delay at local circles. Observations indicate a difference between the acceptance of gaps/lags in the entering and circulating stream of conflicting traffic as well as a difference between critical gaps and critical lags. The mean observed critical gaps/lags are larger than in other countries, which indicates that delays at local circles will be greater. Gap/lag acceptance based on critical distances rather than critical times was applied successfully in the simulation program TRACSIM. A method is proposed to estimate critical distances from the geometric layout of the circle. Critical gaps are not fixed, but should vary with at least the conflicting flows. The investigation of the effect of unbalanced flows on delay, showed that the variability in drivers' critical gaps is more a function of delay than of conflicting flow. Entry delays increase because of an increase in conflicting flows or because of an unfavourable imbalance of conflicting flows. In both instances the drivers' critical gaps will decrease. A variable critical gap model only based on conflicting flows will show no change in the drivers' critical gaps if the conflicting volumes remain constant, even though the actual average delay might increase because of an unfavourable imbalance in conflicting flows. / Thesis (Ph.D.)-University of Natal, Durban, 1997.

Thesis (MScEng)-- Stellenbosch University, 2013. / ENGLISH ABSTRACT: Speed-flow-density relationships are the most useful tools in the highway design and planning process. They are useful in predicting the roadway capacity, in determining the adequate level-of-service of traffic flow and in determining travel time for a given roadway. Two-lane two-way rural highways constitute the vast majority of the rural road network in South Africa. Nowadays in the Western Cape and other provinces of South Africa, the speed-flow-density relationships normally used for rural transportation studies are derived from the Highway Capacity Manual, which reflects the traffic conditions in the North American situation. Since the North American traffic conditions may be different from the South African conditions, a need to investigate speed-flow-density relationships on these highways in South Africa arises in order to justify any investment made on these roads. In this context, a video technique was used to collect traffic flow data during morning peak hours on two rural two-lane two-way highways in the Western Cape Province in order to investigate these relationships. Through the use of Adobe premiere C.S 6 software, travel time of individual vehicles and distance headways were measured and used in computation of average speed and average density. Several researchers have developed models to describe the relationships between traffic characteristics on uninterrupted flow facilities. In this study, some of these models were tested using collected data in order to investigate which model fits the data satisfactorily. Statistical methods were used to evaluate the ability of each model to predict the flow characteristics over the whole range of data. Average speed and density data were used through regression analysis to perform curve fitting and testing of these developed models. In the next stage, the model which provided a best representation of the data on each section was selected and through the application of the steady-state equation (2.1), flow-density and speed-flow relationships were established on these sections. The available data were also used to investigate the impact the observation time has on the speed-flow curve and the resulting capacity value. Finally, the developed speed-flow curves were used to determine the capacities of the study sections. These capacity values were used to determine if the shoulder usage contributes in increasing the capacity of two-lane two-way highways by comparing them to the capacity provided by HCM. / AFRIKAANSE OPSOMMING: Spoed-vloei-digtheid verhoudings is baie handig in die beplanning en ontwerp van paaie. Dit kan ook gebruik word in die voorspelling van kapasiteit, diensvlak en reistyd. Twee-laan twee-rigting paaie maak die grootste deel van die Suid-Afrikaanse padnetwerk uit en vir die beplanning daarvan word van Amerikaanse spoed-vloei-digtheid verhoudings gebruik gemaak aangesien daar nog nie voorheen ‘n studie hiervan in SA gemaak is nie. Video-opnames is gebruik om verkeersvloeidata op twee paaie in die omgewing van Stellenbosch te versamel. Die reistyd en digtheid van individuele voertuie is tydens spitstye waargeneem. Die data is gebruik om te bepaal watter modelle die beste is om die spoed-vloei-digtheid verhoudings vir hierdie paaie te modelleer. Die beste modelle is dan gebruik om die kapasiteit van die paaie te bepaal en dit te vergelyk met die Amerikaanse waardes.