MURP / Department of Urban and Regional Planning / The study investigated commuters’ perceptions towards the implementation of bus rapid transport (BRT) system in Polokwane Local Municipality. This was achieved through determining the commuters’ perceptions of the implementation phases of Polokwane BRT. The assessment considered whether the implementation of the Polokwane BRT meets commuters’ travel needs and preferences with respect to the operational characteristics. The study used the stated preference and checklist technique to collect data. Questionnaires were spatially distributed per settlement cluster, targeting the general commuters. Data was also collected through interviews with the company implementing the system and the Polokwane municipal officials. The study was based on a comparison of two scenarios, which are: “protraction of public transport system without BRT and a combination of both public transport and the BRT system”. The collected data was analysed using Scenario-Based Analytical Approach (SBAA), with the aid of Statistical Package for Social Sciences (SPSS) and Microsoft Excel software. The collected data was represented and interpreted using both quantitative and qualitative data presentations formats. The study noted that commuters’ travel needs were limited to comfort, time, fare (affordability), safety and security. The results revealed that the methods used to introduce the implementation of BRT system influenced the perceptions of commuters towards the system. Most of survey respondents (55%) were not aware of BRT system implementation, while (56%) indicated that the methods used were extremely ineffective. Furthermore 43% of the survey respondents were not willing to shift from the public transport to use the BRT system. Therefore, many of the survey respondents (52%) negatively perceived and associated the system with negative transport changes within the Municipality. However, it was also established that the operation of the system within the municipality would attract (77%) commuters towards public transportation such as BRT system, only if the system incorporates commuters’ preferred operational characteristics. On the other hand, the scenarios revealed that without the intervention of BRT system, the municipality is likely to continue experiencing transport problems, although there are also challenges associated with the implementation of BRT system. The use of effective and familiar methods of communication, integrating the system with existing modes of transport, development densification, provision of mixed land use and transit oriented development along BRT corridors and BRT routes extensions to residential and socio-economic areas, are some of the recommendations in support of the BRT system. / NRF

Diese Diplomarbeit beschäftigt sich mit der "schleichenden Abschaltung" begrenzter Netzzugänge bei Überschreitung eines vorgegebenen Datenvolumens und der Anwendung dieser Technologie im Chemnitzer Studentennetz. Nutzer, die bestimmte Transfervolumina überschreiten, werden schrittweise einer immer schlechter bewerteten Verkehrsklasse zugeordnet, wo ihnen weniger Bandbreite zur Verfügung steht. Dazu werden die Möglichkeiten, die der Linux-Kern hinsichtlich Quality of Service bietet, genutzt.

Gemeinsamer Workshop von Universitaetsrechenzentrum und Professur Rechnernetze und verteilte Systeme der Fakultaet fuer Informatik der TU Chemnitz. Der Dynamic Traffic Shaper erlaubt ein dynamisches Bandbreiten-Management von begrenzten Netzzugängen unter Zuhilfenahme der Quality-of-Service-Möglichkeiten des Linux-Kerns. Er wird typischerweise auf einem Router, der das gemanagete lokale Netz mit dem Internet verbindet, eingesetzt. Nutzer dieses Netzes werden anhand ihrer Transfervolumina für eine gewisse Zeit in verschiedene Nutzergruppen mit unterschiedlichen Bandbreiten eingeordnet. Damit soll sichergestellt werden, dass über einen bestimmten Zeitraum nur ein festgelegtes Datenvolumen verursacht werden kann. In diesem Vortrag werden die aktuellen Entwicklungen beim Einsatz im Chemnitzer Studentennetz aufgezeigt.

Philosophiae Doctor - PhD / Cloud Computing enables users achieve ubiquitous on-demand , and convenient access to a variety of shared computing resources, such as serves network, storage ,applications and more. As a business model, Cloud Computing has been openly welcomed by users and has become one of the research hotspots in the field of information and communication technology. This is because it provides users with on-demand customization and pay-per-use resource acquisition methods.

In this paper, we introduce a framework for the adaptive control of universal mobile telecommunications system (UMTS) networks in order to improve bandwidth utilization of the radio channels. The key contribution of the paper constitutes the introduction of a performance management information base for dynamically adjusting the packet scheduler and admission controller. Thus, the adaptive control framework closes the loop between network operation and network control. Furthermore, the adaptive control framework can effectively deal with the different time scales of packet scheduling and admission control. Moreover, we present a traffic model for non-real-time UMTS traffic based on measured trace data. The analysis and scaling process of the measured trace data with respect to different UMTS bandwidth classes constitutes the basic concept of this traffic characterization. Using this traffic model and simulation on the IP level, the gain of employing the adaptive control framework is illustrated by performance curves for various quality of service measures.

El objetivo del presente trabajo de investigación es determinar los factores más influyentes en la calidad percibida del servicio de taxi asistido por aplicativo móvil, desde la percepción de los usuarios que viven o trabajan en Lima Moderna. Para ello, se utilizó la metodología del modelo Servqual y se realizó un estudio no experimental del tipo descriptivo. El instrumento de recolección fue una encuesta estructurada de 22 preguntas, tanto para las percepciones como para las expectativas. La diferencia entre ambas dio como resultado brechas negativas del servicio. Las preguntas se agruparon en cinco factores (fiabilidad, sensibilidad, seguridad, empatía y tangibilidad). La encuesta se aplicó a 388 usuarios, lo que permitió conocer el nivel de satisfacción por cada factor, por sexo y por empresa. Entre los resultados más relevantes de la investigación se pudo evidenciar que los usuarios se sienten insatisfechos con la calidad del servicio, especialmente los relacionados con los factores de seguridad y empatía. El trabajo también investiga la relación entre cada factor y la calidad percibida, estableciéndose una correlación moderada. Además, se usa la herramienta Net Promoter Score que mide la lealtad de los usuarios basándose en sus recomendaciones. Si bien en el Perú no existen estudios que aplican Servqual para medir la calidad de servicio en la industria del taxi por aplicativo, este es muy usado internacionalmente. Confiamos que esta investigación servirá de base para futuros estudios a nivel nacional. Nuestra investigación puede ser aprovechada por los directivos de las empresas de taxi por aplicativo para realizar acciones de mejora en su servicio. / The aim of this recent investigation is to determine the most significant aspects that influence in the quality of taxi service apps in Modern Lima. For this, it was used the methodology of Servqual Model and it was developed as a non-experimental research of descriptive design. The instrument to collect data was a structured survey of 22 questions for both, perceptions, as well as expectations. The difference between them result in negative gaps in the service. The questions were classified according to five aspects (reliability, responsiveness, assurance, empathy and tangibles). The survey was applied to 388 users, which allowed to know the level of satisfaction in each factor, by gender and company. Among the most relevant results of the investigation it was noticeable that user feel unsatisfied with the quality of the service, especially those related to factors of reliability and empathy. Likewise, the research attempts to stablish the relationship between each factor and the quality perceived, stablishing a moderate correlation. Furthermore, it was used the Net Promoter Score tool, which measures the users’ loyalty based on their recommendations. Although in Peru there has not been any research that apply Servqual to measure the quality of service in transportation network companies, it is a model used internationally. We trust that this research will be used as a reference for future local studies. This investigation can benefit managers of taxi companies in order to carry out improvements in their service. / Tesis

La presente investigación tuvo como principal objetivo determinar la percepción de calidad de las usuarias de los servicios Control Prenatal y Parto en clínicas de categorías II y III de Lima Metropolitana y cómo esta información puede ser considerada como valiosa para generar un incremento de la demanda de atenciones en dichos establecimientos de salud. La metodología empleada se ha enmarcado dentro del enfoque cuantitativo porque ha estado conformada por un conjunto de procesos secuenciales que buscan determinar el nivel de percepción de la calidad. Para analizar las dimensiones que constituyen la calidad del servicio se ha utilizado la metodología Valoración del Desempeño – SERVPERF (por sus siglas en inglés). Para evaluar la percepción de la calidad de servicio se han establecido las siguientes dimensiones: Fiabilidad, Seguridad, Tangibilidad, Capacidad de Respuesta y Empatía. Para ello se construyó un cuestionario, que fue validado por criterio de especialistas y estadísticamente mediante el Coeficiente Alfa de Cronbach; obteniendo un resultado equivalente a 0,935 que indica una alta confiabilidad. El resultado general ha mostrado que sólo el 32,9% de las encuestadas ha percibido la calidad de los servicios de Control Prenatal y Parto en clínicas de categorías II y III de Lima Metropolitana como buena. El 65,3% la ha percibido como regular y un 1,8%, como mala. Las dimensiones más valoradas como buenas han sido las de Seguridad y Tangibilidad con un 42,67% y 40,87%, respectivamente. La Capacidad de respuesta ha sido la dimensión más valorada como mala (11,31% del total de las personas encuestadas). / The main objective of this research was to determine the perception of quality of the users of prenatal control and childbirth services in category II and III clinics of Lima Metropolitana and how this information can be considered valuable to increase in the demand for care in these health facilities. The methodology used has been framed within the quantitative approach because it has been made up of a set of sequential processes that seek to determine the level of perception of quality. The Performance Valuation – SERVPERF methodology has been used to analyze the dimensions of service quality. The following dimensions have been established to assess the perception of quality of service: Reliability, Safety, Tangibility, Responsiveness and Empathy. To do this, a questionnaire was built, which was validated by specialist criteria and statistically by the Cronbach Alpha Coefficient; obtaining a result equivalent to 0.935 indicating high reliability. The overall result has shown that only 32.9% of respondents have perceived the quality of Prenatal Control and Childbirth services in category II and III clinics in Lima Metropolitana as good. 65.3% have perceived it as regular and 1.8% as bad. The dimensions most valued as good have been Safety and Tangibility with 42.67% and 40.87%, respectively. Responsiveness has been the most valued dimension as bad (11.31% of the total respondents). / Tesis

Service assurance is critical for high-demand services running on telecom clouds. While service performance metrics may not always be available in real time to telecom operators or service providers, service performance prediction becomes an important building block for such a system. However, it is generally hard to achieve.  In this master thesis, we propose a machine-learning based method that enables performance prediction for services running in virtualized environments with Docker containers. This method is service agnostic and the prediction models built by this method use only device statistics collected from the server machine and from the containers hosted on it to predict the values of the service-level metrics experienced on the client side.  The evaluation results from the testbed, which runs a Video-on-Demand service using containerized servers, show that such a method can accurately predict different service-level metrics under various scenarios and, by applying suitable preprocessing techniques, the performance of the prediction models can be further improved.  In this thesis, we also show the design of a proof-of-concept of a Real-Time Analytics Engine that uses online learning methods to predict the service-level metrics in real time in a container-based environment.

For an IT company that provides a service over the Internet like Facebook or Spotify, it is very important to provide a high quality of service; however, predicting the quality of service is generally a hard task. The goal of this thesis is to investigate whether an approach that makes use of statistical learning to predict the quality of service can obtain accurate predictions for a Voldemort key-value store [1] in presence of dynamic load patterns and network statistics. The approach follows the idea that the service-level metrics associated with the quality of service can be estimated from serverside statistical observations, like device and network statistics. The advantage of the approach analysed in this thesis is that it can virtually work with any kind of service, since it is based only on device and network statistics, which are unaware of the type of service provided. The approach is structured as follows. During the service operations, a large amount of device statistics from the Linux kernel of the operating system (e.g. cpu usage level, disk activity, interrupts rate) and some basic end-to-end network statistics (e.g. average round-trip-time, packet loss rate) are periodically collected on the service platform. At the same time, some service-level metrics (e.g. average reading time, average writing time, etc.) are collected on the client machine as indicators of the store’s quality of service. To emulate network statistics, such as dynamic delay and packet loss, all the traffic is redirected to flow through a network emulator. Then, different types of statistical learning methods, based on linear and tree-based regression algorithms, are applied to the data collections to obtain a learning model able to accurately predict the service-level metrics from the device and network statistics. The results, obtained for different traffic scenarios and configurations, show that the thesis’ approach can find learning models that can accurately predict the service-level metrics for a single-node store with error rates lower than 20% (NMAE), even in presence of network impairments.

With its enormous potential Voice over Internet Protocol is one of the latest buzzwords in information technology. Despite the numerous advantages of Voice over IP, it is a major technical challenge to achieve a similar call quality as experienced in the ordinary Public Switched Telephone Network. This thesis introduces standardized Internet protocols for Voice over IP, such as Session Initiation Protocol (SIP), Real-time Transport Protocol (RTP), in its background chapter. In order to provide better Quality of Service (QoS) Voice over IP applications should support a feedback mechanism, such as the Real-time Control Protocol (RTCP), and use accurate timing information, provided by the Network Time Protocol (NTP). Additionally this thesis considers synchronization issues in calls with two and more peers. After a rather academic overview of Voice over IP, the open source real-time application “minisip”, a SIP user agent, and its operation and structure for handling audio streams will be introduced. Minisip was extended by an implementation of NTP and RTCP to provide a test platform for this thesis. A clear conclusion is that the addition of global time helps facilitate synchronization of multiple streams from clients located any where in the network and in addition the ability to make one-way delay measurements helps SIP user agents to provide better quality audio to their users. / Röst över IP, eller Internettelefoni baserad på “Voice over Internet Protocol” (VoIP), har med sin stora potential blivit ett av de senaste modeorden inom informationsteknologin. Vid sedan av ett antal fördelar med VoIP så innebär det en stor teknisk utmaning att uppnå en likadan samtalskvalitet som i det vanliga, fasta, telenätet. I den här uppsatsen beskrivs hur tjänstevalitet för VoIP kan förbättras genom att noggrant tidssynkronisera de (två eller flera) klienter som deltar i ett telefonsamtal. För detta krävs dels en återkopplingsmekanism, såsom “Real-time Control Protocol” (RTCP), samt en gemensam tidsuppfattning i de inblandade klienterna, vilket kan uppnås med hjälp av “Network Time Protocol” (NTP). Dessa protokoll, liksom de övriga Internet-standarder som VoIP baseras på (såsom “Session Initiation Protocol” (SIP) och “Real-time Transport Protocol” (RTP), beskrivs inledningsvis i uppsatsen. För studien har en SIP-klient baserad på öppen källkod använts (“Minisip”), och utökats med NTP och RCTP funktionalitet för att testa den föreslagna förbättringen av VoIP. En tydlig slutsats är att kännedom om en “global tid” möjliggör synkronisering av multipla ljudströmmar från klienter som befinner sig på olika nätverk. Möjligheten att mäta paketfördröjningen (envägs) bidrar också till en förbättrad ljudkvalitet.