Flood Monitoring & Forecasting using SAP HANA
World Metrological Organization reports that out of majority of all disasters in the world, flooding is one of the most severe disasters affecting the people across the globe.
India is one of the worst flood affected country in the world and total Indian rainfall is concentrated over a short monsoon season of four months. As a result, the rivers witness a heavy discharge during these months, leading to widespread floods.
Several states suffer from severe annual flooding. Unfortunately, they do not have an early warning mechanism that would alert the affected regions from the occurrence of a disaster. The existing disaster management mechanism is primarily focused on strengthening rescue and relief arrangements during and after disasters. As a result:
- Irrigation department still has a manual and slow process to collect and consolidate the data on water flow, reservoir levels and weather across various regions
- There is no computational and data processing capability for flood forecasting calculations
- Absence of automated mechanism to alert on the flood situations leads to widespread loss of life and property
It is desired to have a technological solution that could
- Process large data (structured) volumes coming from sensors installed across river basins and SMS text messages ( unstructured ) from citizens
- Provide analytical capability to detect and forecast the flood situation.
- Provide dashboard application to be consumed in a command center (large TV screens ) and Tablets ( for chief engineers of irrigation department ).
SAP HANA platform with its high performance, advanced analytical capabilities (Predictive, Geospatial, Text analytics) and UI5 technology front end is suited well to address these technical challenges.
A solution, built on SAP HANA, would enable the state irrigation department to centrally monitor, analyse and forecast flood situations. It can automatically alert the administration and the population about probable occurrence of a flood event. This would help take necessary measures to minimize the loss of human lives and mitigate the damage to properties.
Arteria Flood Real-time Monitoring (FoRM), uses sensor data from various sites such as reservoirs, dams, canals and consolidates it using the SAP HANA platform for monitoring in real time via a central dashboard using maps. Currently, sensor data from 4 major river basins across roughly 1000 sensor points is collected into SCADA servers and pulled into HANA platform every 5 minutes. Automatic alert notifications are calculated in case of a threshold breach and predictive algorithms are used to forecast a flood situation based parameters such as rainfall, water level and discharge levels.
FoRM has the potential to save the loss of property and lives of citizens by accurate forecasts of flood situations and the impacted geographic zones.
Nice blog....integration of HANA & SCADA
Very interesting app!
I'm curious how you get data into HANA.
SLT, Data Services, HANA smart data integration/streaming/access, something else?
At the moment Data Services for the pilot. Would keep this comment updated as we move towards implementation and might change the technology.
Thanks for this nice and useful blog.
Do use use PAL or R for your predictive analysis algorithm?
Thanks Andre. The first phase / pilot focused on monitoring and predictive part will come in later as we progress.