Návrh a implementace řešení reportingu pro management oddělení IT společnosti

Abstract

The Master´s thesis entitled, “Design and Implementation of a Reporting Solution for the Management of an IT Company Department”, is focused on analysis of the selected IT Company, in collaboration with its processes, user requirements and design and implementation of reporting platform for reports and dashboards processing. The aim of reporting platform is to provide results department more clearly. The purpose of the platform is to increase the efficiency of the processes in the department and ensure better control over the quality of service provided, by standardization KPI, metrics, reports, the process of their creation, multidimensional view of data and interactive reporting. The first part of Master´s thesis is focused on the description of important theoretical knowledge which are vital for drawing up the practical part. The thesis deals with terms such as Business Intelligence, data modeling, Kimball’s Dimensional Model Design Life Cycle , multidimensional analysis techniques and dashboards design on the principle of User Experience. The second part is devoted to a description of the selected company including organizational structures and processes. In-depth description is focused on IT department itself, together with processes for which are designed dimensional model. There are listed current reports and metrics due to their principles and limitations. There are charted roles of reporting, used tools, processing and human resources, sources for reports and the possibility of storing and sharing. The third part is divided into dimensional modeling, implementation data pump routines and design of dashboards. In the first step of dimensional modeling there are specified all processes of the company, evaluated by various kinds of factors, and it is been set a priority for determining the order of processing. Moreover there are identified the source data systems which play an important part to measure the performance and quality of defined process. According to the selected methods there are finally collected and evaluated all the requirements of users. The second step set the granularity of fact table and it determines preliminary dimensions and facts. In the third step the dimensions are found including their attributes (key and non-key attributes), the target resources and the rules for data pump. Numerical facts are specified in the identification phase of the facts. On the basis of these four steps it is been constructed a conceptual and a logical model. The physical design is primarily focused on aggregation, indexing and partitioning. Then it is drawn up a physical model. Taking implementation data pump routines into consideration it is necessary specified source and target data structures. Thereafter implemented data pump routines are designed, to load data into the dimensional model. For the design output dashboards on the principle of User Experience, there are first collected the requirements of users and it is constructed rough design of outputs. In the final phase dashboards are physically designed, it is performed a résumé of the requirements from the first step of dimensional modeling and it is tested the functionality of dashboard.