Tutorial: AI for Data-Driven Decisions in Water Management, in conjunction with Association for the Advancement of Artificial Intelligence (AAAI), at San Francisco, USA, on Feb 04, 2017


Date of tutorial is Feb 04, 2017 from 2:00-3:45 PM. See tutorial schedule here.

Intended Audience

The intended audience is a beginner to medium experienced researcher who is curious about AI in general and sustainability applications in particular. The tutorial will expect the audience to have a basic Computer Science background consisting of data structures, algorithms, databases and web applications.

Related Tutorials

1. AAAI 2015 tutorial, AI for Smarter Cities. Hype or reality? A Study in Dublin, Bologna, Miami and Rio.

2. IJCAI 2015, AI for Smart City Innovations with Open Data, at Buenos Aires, Argentina, July 25-31, 2015


[8 Feb 2017] See blog summarizing the tutorial.
[28 Jan 2017] Slides posted on slideshare.
[05 Jan 2017] Tutorial site is up.


  1. Presentation: Main deck - on slideshare (backup: on drive).
  2. Demo: BlueWater site and GangaWatch app (download directly from .apk or from Google playstore - India region only; blog).
  3. Demo: NECTAR, Nirikshana (Hindi word for Inspection) for Enforcing Compliance for Toxic wastewater Abatement and Reduction
  4. Resource: API details and sample code. See Bluewater's data page here.


Water is unique in its role as a life preserver. It is important to all members of a society. Abhay may want to take a bath in the river during a religious festival and would want to know which banks of the river are feasible to go without getting sick. Bina may want to fetch water for household activities. Chetan may want to use river water for irrigating his fields. Divya may wonder if fishing or vegetable growing is promising on the river catchment area to supplement her family's earnings. Eashwar may want to help river dolphins increase in strength so that more tourists can come to his river-side hotel. Farida, a doctor, may want to warn patients about water-borne diseases so that diseases in the area come down.

They and many more can benefit with decision support aids (i.e., AI systems) which can help them understand water pollution data and alternative decision choices that they may have. However, if one is looking for quality pollution data, one is lost. This is surprising given that there is a rich history in many countries of field-visits for data collection and lab-based testing, and they look forward to adopting real-time water sensing in a big way.

In this tutorial, we want to educate AI researchers about data issues in water management and equip them with resources so that they can employ their techniques to pressing problems that impact everyday lives. We ground the presentation based on our work with (a) open data released by government agencies, (b) developing the BlueWater platform for quantitative pollution data, and (c) releasing the NeerBandhu (WaterFriend) mobile app to collect qualitative data from people. Open data is available as-is but not familiar to many in research community. BlueWater is a data platform [2a-2c] to collect, reconcile and share quantitative pollution data from real-time sensors and historical data, and powers the released GangaWatch mobile app to make sense from all. Since mobile phones are omni-present, mobile apps like CreekWatch on iOS and Neer Bandhu (Water Friend) on Android enable people to share qualitative assessment of water quality they interact with. We have used them on field experiments at polluted rivers in India like Yamuna at Delhi, Hindon at Meerut and Ganga at Haridwar; generated novel insights for environment use-cases, and released collected data via APIs to community.


The topics we will cover are:

  1. Water management problem
    1. Water supply and demand
    2. Pollution basics – pollutants and their impact
    3. Use-cases where pollution data make a difference
  2. Water pollution data
    1. Water pollution as Open data
    2. Quantitative data collection

                                               i.     Real-time sensors

                                              ii.     Sensor deployment – fixed and non-stationary sensors

                                             iii.     Correlation with lab-tests

                                             iv.     Publishing geo-tagged data

                                              v.     BlueWater and GangaWatch experience in Indian setting

    1. Qualitative data collection via mobile apps

                                               i.     Pros and cons of qualitative data

                                              ii.     Publishing geo-tagged data

                                             iii.     NeerBandhu experience in Indian setting

    1. Working with quantitative and qualitative data together

                                               i.     Geo-spatial correlation

                                              ii.     Data quality assessment

    1. Visualization
  1. End-user analytics
    1. Understanding geo-spatial impact of large-scale tourist movement near a river
    2. Inspection plan generation for catching polluting industries
    3. Optimization of sewage treatment
  2. Supporting topics
    1. Anatomy of illustrative apps and platform: CreekWatch / NeerBandhu (Water Friend); GangaWatch; BlueWater
    2. Privacy considerations
    3. Relevant water and data standards 

The AI topics we will cover are:

  1. Data integration
  2. Semantics, knowledge representation and ontologies
  3. Machine learning
  4. Planning, plan management
  5. Game theory, agents
  6. Trust and reputation
  7. Optimization


 The tutorial will run for quarter of a day (1 hour, 45 mins).


Biplav Srivastava, IBM ResearchEmail: biplavs AT us.ibm.com
Sandeep S Sandha, UCLA and ex-IBM Research