The Role of SAP Analytics Cloud Smart Predict Time Series Feature in Evaluating Energy Consumption Trends and Environmental Impact of the SAP Metronom Business Center Office Location

Vera Vavan

Bachelor's thesis

Abstract:

The research constructed for this thesis aims to analyse the energy consumption and the corresponding CO2 production trends and behaviours in one of the SAP Prague office locations, the Metronom Business Center. The date range includes the period from January 1, 2019, until March 31, 2022, and the analysis is performed using the SAP Analytics Cloud solution. The purpose of the analysis is to investigate the changes in office building consumption influenced by the COVID-19 disease emergence, as well as the possible future trends, and whether the solution enables corporate data democratization and to what extent. Analysed data includes the cooling, heating, electricity and water consumption in the office building, from which the corresponding CO2 productions were calculated using emission factors, and the tools from the SAP Analytics Cloud solution used for analysis are the Modeler for creating the data model, Story for data visualizations and Predictive Scenario for time series forecasting. The findings of the data analysis are that the trend of energy consumption has been mostly negative since 2020, with a slight increase in 2021 as a result of office reopening after the COVID-19 shutdown and the catch—up effect. However, the overall decrease in CO2 production is aligned with the SAP’s commitment to reaching carbon neutrality in 2023, since the forecasted trends imply a steady decrease. The solution enablement empowers data democratization due to the simplicity and automation of the tools, nevertheless, the implied limitations include poor personalization options and low confidence of the results. Research restrictions include missing data for electricity consumption for 2019 and 2020, and a restricted number of variables which influence CO2 production due to data confidentiality. Further implementations and suggestions for the Smart Predict feature include building up on the textual explanations and data insights of the results and forecasts, as well as improving pattern recognition.

Keywords

SAC, Predictive model, Forecast, CO2 production, Office building, Data democratization, COVID-19

Language used: English

Date on which the thesis was submitted / produced: 2. 4. 2023

Thesis defence

• Date of defence: 22. 5. 2023
• Supervisor: prof. Ing. Martin Hanel, Ph.D.
• Reader: Martin Vokoun, Ph.D.

Czech University of Life Sciences Prague

Faculty of Environmental Sciences

Bachelor programme / specializace:
Environmental Data Science / Informatics