Empresa Municipal de Aguas de Málaga S.A. (EMASA), the Municipal Water company of Málaga, is a water supply company in the region of Málaga, proving water and services to ca. 600,000 people of the 5th largest city in Spain. EMASA’s goals are to guarantee the sustainability of the water service provided and, at the same time, to improve the level of customer service.

• Its aim was to implement smart metering, in order to alert its customers immediately about internal leaks and encourage responsible consumption of water. 
• EMASA required a ‘big data’ tool to monitor and analyze large volumes of information from its water meters of different meter types and with various frequencies of reading. 
• The Tool needed to have the capabilities to: Find anomalies such as internal leaks in indoor installations, determine which meters should be replaced, identify over- and undersized meters, locate meters that were measuring incorrectly, detect fraud & use technology to provide early warning of incidents in social emergencies

SIWA Meter Data Analytics (MDA)

• SIWA Meter Data Analytics was implemented in EMASA’s water network in Málaga. Data from over 238,000 water meters was normalized and integrated into the database. 
• Currently, 54,400 of the 238,000 meters send daily data readings, including extended information such as battery level, number of starts, sleep time, etc. 
• Thirty parameters were incorporated into the data for each meter (battery level, consumption, average flow, sleep/drip time, continuous consumption, absence of consumption, etc.). These parameters provide the input data to generate alerts in the event of a suspected anomaly or ‘suspicion’. 
• EMASA has identified 64 different kinds of ‘suspicion’ (situations where meters display abnormal behavior), grouped by types (meter replacement, anomalous consumption, over-/under-sized meters, leaks, fraud, administrative fraud, totalizer balance, social emergency). 
• Each kind of ‘suspicion’ takes into account at least one, and potentially many, parameters and is calculated with reference to a pre-determined formula. For example, an absence of consumption generates a social emergency ‘suspicion’ if the inhabitant is a dependent person. In this scenario, family or social services are immediately informed. 
• Fraudulent behavior is another example which generates a ‘suspicion’. Many possible scenarios can lead to a ’suspicion’ of fraud and if/when a potential fraud is identified, the utility can rapidly send a team to conduct an on-site inspection, to confirm if the fraud exists and take quick corrective actions, saving both water and money.