A Real-Time Data Warehouse Solution for Analysis on Indoor Tracking Data

Hansen, Jonas ; Jørgensen, Stig

4. term

Software, Master

2009

66

Dette projekt udvikler et realtids-datawarehouse til at analysere Bluetooth-baserede trackingdata fra Copenhagen Airports A/S i samarbejde med BLIP Systems A/S, som har et tracking-system installeret i lufthavnen. Vi beskriver først, hvordan BLIP Systems lagrer de indsamlede trackingdata, og designer derefter et datawarehouse – et centralt system til at organisere og analysere store datamængder – der kan besvare business intelligence-relaterede spørgsmål (datadrevne indsigter til drift og beslutninger). Som led i dataenes bevægelse fra indsamling til analyse udfører vi datarensning for at rette fejl og uoverensstemmelser. Derudover introducerer vi to casespecifikke algoritmer, ValidMoves og BounceDetection, som hjælper med at normalisere trackingdata, så de er konsistente og klar til analyse. Med specialudviklede applikationer og Microsoft Analysis Server adresserer vi udvalgte spørgsmål og præsenterer resultaterne via databasevisninger og grafiske brugergrænseflader (GUI’er). Vi afrunder projektet med konklusioner og forslag til videre arbejde. En kort sammenfatning findes i Appendix C.

This thesis builds a real-time data warehouse to analyze Bluetooth-based tracking data from Copenhagen Airports A/S, in collaboration with BLIP Systems A/S, which has the tracking system installed at the airport. We first describe how BLIP Systems stores the collected tracking data, then design a data warehouse—a central system for organizing and analyzing large datasets—that can answer business intelligence-related questions (data-driven insights to support operations and decisions). As the data moves from capture to analysis, we perform data cleansing to fix errors and inconsistencies. We also introduce two case-specific algorithms, ValidMoves and BounceDetection, which help normalize the tracking data so it is consistent and ready for analysis. Using custom applications together with Microsoft Analysis Server, we address selected questions and present the results through database views and graphical user interfaces (GUIs). We conclude the project and suggest directions for future work. A brief summary is provided in Appendix C.

[This abstract was generated with the help of AI]

Real-Time Data Warehouse ; Data Warehouse ; Real-Time ; CTF

Download
View record in AAU Student Projects