The thesis is focused on product development in companies in the telecommunications industry. These companies provide products to end customers. The introductory section lists and describes the basic concepts accompanying the issue. Also described herein is a method for achieving the objectives of the thesis, that is based on personal interviews with specialists in selected telecommunications companies. The first objective is to identify variables affecting success launching a new product to market services. The thesis defines the internal and external factors in the organization. Part of the analysis is to assess the current state of internal RFC, that was unexpectedly during the development of a long implementation period (TTM). The content of this thesis is identification of key factors and their impacts on implementation also unveiling constraints of the development process. These constraints are in the view of issues that prevent better performance of the relevant processes. The third and the main objective of this thesis is based on previous analyses, proposal of improvements that could contribute in the future to better product development. The proposed improvements are consulted with a specialist Demand Management to determine the feasibility and potential benefit in practice. The greatest contribution of the thesis lies in the proposed improvements and their assessment for future implementation.

This document specifies a simple signalling mechanism for inter-domain QoS marking using a BGP Extended Community QoS Attribute. Class based packet forwarding for delay and loss critical services is currently performed in an individual AS internal manner. The new QoS marking attribute makes the QoS class setup within the IP prefix advertising AS known to all access and transit ASes. This enables individual (re-)marking and forwarding treatment adaptation to the original QoS class setup of the respective IP prefix. The attribute provides the means to signal QoS markings on different layers, which are linked together in QoS class sets. It provides inter-domain and cross-layer insight into the QoS class mapping of the source AS with minimal signalling traffic.

The global tech company Ericsson has been tracking the return rate of their products for over 30 years, using it as a key performance indicator (KPI). These KPIs play a critical role in making sound business decisions, identifying areas for improvement, and planning. To enhance the customer experience, the company highly values the ability to predict the number of returns in advance each month. However, predicting returns is a complex problem affected by multiple factors that determine when radios are returned. Analysts at the company have observed indications of a potential correlation between alarm data and the number of returns. This paper aims to address the need for better prediction models to improve return rate forecasting for radios, utilizing alarm data. The alarm data, which is stored in an internal database, includes logs of activated alarms at various sites, along with technical and logistical information about the products, as well as the historical records of returns. The problem is approached as a classification task, where radios are classified as either "return" or "no return" for a specific month, using the alarm dataset as input. However, due to the significantly smaller number of returned radios compared to the distributed ones, the dataset suffers from a heavy class imbalance. The imbalance class problem has garnered considerable attention in the field of machine learning in recent years, as traditional classification models struggle to identify patterns in the minority class of imbalanced datasets. Therefore, a specific method that addresses the imbalanced class problem was required to construct an effective prediction model for returns. Therefore, this paper has adopted a systematic approach inspired by similar problems. It applies the feature selection methods LASSO and Boruta, along with the resampling technique SMOTE, and evaluates various classifiers including the Support vector machine (SVM), Random Forest classifier (RFC), Decision tree (DT), and a Neural network (NN) with weights to identify the best-performing model. As accuracy is not suitable as an evaluation metric for imbalanced datasets, the AUC and AUPRC values were calculated for all models to assess the impact of feature selection, weights, resampling techniques, and the choice of classifier. The best model was determined to be the NN with weights, achieving a median AUC value of 0.93 and a median AUPRC value of 0.043. Likewise, both the LASSO+SVM+SMOTE and LASSO+RFC+SMOTE models demonstrated similar performance with median AUC values of 0.92 and 0.93, and median AUPRC values of 0.038 and 0.041, respectively. The baseline for the AUPRC value for this data set was 0.005. Furthermore, the results indicated that resampling techniques are necessary for successful classification of the minority class. Thorough pre-processing and a balanced split between the test and training sets are crucial before applying resampling, as this technique is sensitive to noisy data. While feature selection improved performance to some extent, it could also lead to unreadable results due to noise. The choice of classifier did not have an equal impact on model performance compared to the effects of resampling and feature selection.

Inhalation risks on human health for hazardous air pollutants emitted from MACT I petroleum refining industry were determined using EPA HEM-3 Program. Methodology included compiling vertical and fugitive emissions from 2002 National Emissions Inventory for sources inside two facilities in Louisiana, 'Motiva Norco' and 'Valero St. Charles' refineries. Six cases were modeled applying EPA criteria, where cancer risks are 'low' if the probability is. 1/1, 000, 000, and non-cancer risks are harmful when hazard quotient is > 1. It was demonstrated that fugitive emissions have more impact on human health than the verticals because of their significant portion of the total refining emissions. HAPs can cause moderate adverse effects in humans living nearby refineries, as 113 people resulted in high risk of respiratory problems with Valero emissions, 4571 people resulted in 'moderate' risk of getting cancer with Motiva emissions, 2702 people with Valero emissions, and 11, 282 people with both refineries' emissions.

In the most of the drilling operations it is obvious that a considerable amount of money is spent for drilling related problems / including stuck pipe, lost circulation, and excessive mud cost. In order to decrease the percentage of non-productive time (NPT) caused by these kind of problems, the aim is to control annular frictional pressure losses especially in the fields where pore pressure and fracture pressure gradient is too close which is called narrow drilling window. If we can solve these problems, the budget spent for drilling the wells will fall, therefore enabling the industry to be able to drill wells that were previously uneconomical. Managed Pressure Drilling (MPD) is a new technology that allows us to overcome these kinds of drilling problems by controlling the annular frictional pressure losses. As the industry remains relatively unaware of the full spectrum of benefits, this thesis involves the techniques used in Managed Pressure Drilling with an emphasis upon revealing several of its lesser known and therefore less appreciated applications.

The aim of this thesis was studying and describing known methods for testing transmission parameters in data networks based on protocol stack TCP/IP. Firstly were individual transmission parameters specified and determined their impact on service quality. Another aim was to carry out a comparison of the different methodologies of measurement of transmission parameters, performed their qualifications on the basis of these findings and methodology for measuring service quality from the user's perspective was proposed. It was a web application where users measured basic transmission parameters, then the users judged videos according to the evaluation scale MOS (Mean Opinion Score) and with the help of these data was analysis, classification connection and impact of connection on MOS evaluation.

La dérégulation des marchés électriques modifie en profondeur l'approche technico-économique dans l'exploitation et l'optimisation des réseaux électriques. C'est dans ce nouveau contexte que les spécialistes des réseaux électriques se voient confrontés à de nombreux défis. Ainsi, les grands réseaux électriques sont utilisés de plus en plus près de leur limite de stabilité. Cependant, faire fonctionner un réseau trop près de ses limites créer des oscillations de puissance entre les centrales de différentes régions des réseaux. Ces oscillations sont appelées oscillations <

<p>It is very probable that VoIP will soon replace the ordinary telephone. Beside all advantages of the digital voice-connection it is linked to the danger of spam on the telephone. A lot of approaches have been developed to solve the problem of VoIP spam. Because some of these solutions are based on access to personal information of its users, a broad discussion about the best and most ethical approach has started.</p><p>This thesis analyzes the users’ point of view towards the VoIP spam problem and the extent of users’ willingness to offer private information in order to avoid VoIP spam. It presents results from a qualitative and a quantitative research as well as approaches for a most realistic- and most promising VoIP solution. These new approaches are based on the results of the research.</p><p>The main points of the results showed that users were not willing to offer private information to companies and that they were not willing to pay any amount of money for VoIP spam solutions. Users held governmental organisations and telephone operators responsible for finding a solution against VoIP spam.</p>

It is very probable that VoIP will soon replace the ordinary telephone. Beside all advantages of the digital voice-connection it is linked to the danger of spam on the telephone. A lot of approaches have been developed to solve the problem of VoIP spam. Because some of these solutions are based on access to personal information of its users, a broad discussion about the best and most ethical approach has started. This thesis analyzes the users’ point of view towards the VoIP spam problem and the extent of users’ willingness to offer private information in order to avoid VoIP spam. It presents results from a qualitative and a quantitative research as well as approaches for a most realistic- and most promising VoIP solution. These new approaches are based on the results of the research. The main points of the results showed that users were not willing to offer private information to companies and that they were not willing to pay any amount of money for VoIP spam solutions. Users held governmental organisations and telephone operators responsible for finding a solution against VoIP spam.

The target of this master's thesis were the known methods for testing the quality of transfer parameters of data networks. I have studied the RFC 2544 document to analyze these tests. From that information and by studying other existing web-based applications I was able to design a concept of my own application that will allow to measure basic quality parameters of transfers made through the Internet (the parameters: downstream, upstream, latency, variance of these and a traced route). My application doesn't require any modifications on the user's system. It's built on the server-based programming language PHP and uses the relational database engine MySQL to store measurement and user data. On the client side, it's assisted by the JavaScript scripting language. Both registered users and visitors are allowed to perform the listed measurements. Registered users are able to browse the history of their own benchmark results and also to send messages to others. There is an administrative account to oversee the operations. Another part of my thesis work is an application for Windows that performs the same measurements but does not use JavaScript.