Maintaining a safe, sustainable water supply isn’t just about drilling a well or installing a pump: communities need to be able to manage and maintain the infrastructure for themselves.
Water Mission is working with IBM to capture data from each of its projects and transform it into actionable insight that will flow back down to each community, helping them make smarter decisions.
Empowerscommunities to maintain water supplies that meet rigorous safety standards
Revealsconsumption patterns to help develop innovative sustainable business models
Unlocksthe ability to predict issues before they occur and take rapid preventive action
Business challenge story
Tackling a global water crisis
Around the world, an estimated 1.8 billion people lack access to safe water. In addition, 2.4 billion people lack basic sanitation, and 842,000 people die every year from water-related diseases. Water security is also a huge issue: a large proportion of the world’s population lack the capacity to safeguard their access to water supplies, and droughts and floods alone account for 70 percent of all economic losses in sub-Saharan Africa.
Water Mission was founded in 2001 as a direct result of its founders’ experience in responding to a natural disaster: Hurricane Mitch, which hit Honduras in 1998. Confronted by the realization that billions of people are forced to drink dirty water every day, Molly and George Greene III decided to start a non-profit organization that would work to provide clean, safe and sustainable water supplies for as many people as possible worldwide.
George Greene IV, President and COO of Water Mission, explains the organization’s vision: “Access to water is important, but access alone is not enough. It’s also vital to ensure that the water supply is safe and supporting management systems are sustainable—and for communities in some of the world’s poorest countries, that can be a major challenge.
“Creating a sustainable water supply isn’t just a technical issue, it involves understanding culture, governance, affordability and a host of other factors. Our projects work with communities to help them transform the way they manage their water supplies, so it’s as much about education as it is about infrastructure.”
Sandy Funk, Director of Marketing and Communication, adds: “For communities to ensure that they can be self-sufficient for years to come, they need to be able to support and maintain their water supply both operationally and financially. A lot of what we focus on is helping them create business models that will give everyone access to safe water, while also building up funds to maintain and invest in the infrastructure over the long term.”
Seth Womble, Vice President of Operations, comments: “Different communities have different needs, so there’s no single model that will work in every situation. For example, in some cases, you can power a water pump from the grid; in others, you need to look at different options such as solar-powered pumps or fuel-powered generators. And from a financial perspective, different business models can be successful depending on the community’s resources and needs.
“Nevertheless, there are still best practices that you can identify, and we are always looking to learn more about what works and what doesn’t. Over the past 15 years we’ve explored a huge number of options in different projects, and collected an enormous amount of data—ranging from engineering reports to financial transactions, and even sensors embedded in the water treatment stations themselves. We knew that this data offered a potential goldmine for our researchers and project coordinators—we just needed a little help to analyze these massive amounts of data and turn them into something actionable.”
Helping the data flow
Water Mission decided to work with IBM® jStart®, an organization that champions rapid transformation for clients on the cognitive journey. The jStart team has been identifying, extending and applying breakthrough technology for 20 years, and is now helping Water Mission to explore the possibilities of harnessing cutting-edge analytics technologies to unlock the value of its unique datasets.
George Greene IV comments: “Over the past two years we’ve made significant investments in growing our IT organization and upgrading our capabilities, and we have made good progress with our ERP and CRM systems. However, these are off-the-shelf systems, and their requirements are relatively straightforward to manage.
“We had also developed a homegrown system for supporting our program activities. We refer to it as the PUMP (Program Update and Management Platform), and we use it to manage all our project-related information. The volume and variety of data in the PUMP was so great that we were struggling to analyze it effectively.
“That’s where jStart came in—they brought the expertise to help us apply cutting-edge analytics techniques and technologies to help us get ahead of the curve, and transform our data into actionable insight to help communities around the world.”
Kristen Check, Program Evaluation Coordinator, recalls: “One of the first things we did with IBM was to look at data from three pilot projects in Uganda, where people pay for water by loading money onto a prepaid card at a kiosk, and then scan that card at the water dispenser to collect their water. This data is available to us via an online management platform. The jStart team helped us drill down into more than a year of historical water consumption data, and understand water purchases at an individual level. This gave us a much deeper level of insight than we ever had before, and revealed some very useful findings.”
The jStart team loaded the data into IBM Analytics for Apache Spark, a managed service that provides the lightning-fast performance data scientists need to answer complex big data questions quickly. To make it easy to explore the data, create statistical analyses and present the results, they connected the Spark engine to a Jupyter Notebook front-end. The combination of Spark and the Jupyter Notebook made it easy to perform rapid, iterative analysis and reach accurate conclusions quickly.
In addition to this type of specific analysis, the jStart team is also helping Water Mission develop tools that it will be able to use to analyze and monitor data on an ongoing basis—and even build predictive models to help it optimize operations.
George Greene IV explains: “We have a host of different monitoring systems and ways of gathering data. Some of it is collected manually by technicians out in the field, or by members of the community. In other cases, we’re using a low-cost satellite modem system that gathers data from water flow sensors and uploads it to the cloud every three days to minimize data transmission costs. Ideally, if we can find a way to make it sustainable, we would like to include some degree of sensor-based monitoring for all our projects going forward.”
Ryan Reif, Information Systems Manager, adds: “The jStart team is helping us build a dashboard that aggregates and summarizes all of this data, and automatically sends email alerts to the in-country managers about important events or trends. Instead of having to look through the data ourselves to see if anything is out of the ordinary, the system notifies us immediately. By making it easier to identify patterns in the data, the system helps us gain a better overview of the situation at each of our projects, and identify opportunities to improve.”
The dashboarding and alerts system is built on IBM Compose for MongoDB, a fully managed cloud database service that allows Water Mission to store its sensor data, engineering reports and other data in a JSON document store.
Unlike a traditional relational database platform, Compose for MongoDB can store data without converting it into a rigid table structure, which makes it especially suitable for ingesting data from Internet of Things devices such as Water Mission’s water flow and water quality sensors. As the range and types of sensors deployed in the field continue to expand and evolve, it is important to be able to capture and store the data they produce in a flexible way. MongoDB’s schemaless architecture and JSON data structure are a perfect fit with this requirement.
In the future, Water Mission and jStart plan to take the monitoring capabilities to the next level by developing predictive models. By analyzing trends in the data, these models will help Water Mission identify potential challenges and opportunities before they arise, and take appropriate action quickly.
“Many of our projects are in remote areas, and it can take days for one of our project managers or technicians to travel there,” says Sandy Funk. “That’s why predictive analytics could be such an advantage for us: it could help us identify what a community is going to need before they actually need it. That means we can make sure we send the right people and the right resources at the right time.”
Tapping into a clear stream of insight
IBM jStart is helping Water Mission gain deeper insight into its operations, which will help it empower more communities to build reliable, safe and sustainable water supplies. By using analytics to identify successful operational, financial and cultural models, Water Mission also hopes to be able to share best practices with other organizations, such as UNICEF and UNHCR, and work together with them to make a significant impact on the lives of the 1.8 billion people who currently lack a safe water supply.
Kristen Check comments: “We’re looking to put what we’ve learned during our work with IBM into practice straight away. For example, when we analyzed the data from our projects in Uganda, we realized that some of the prepaid water cards—often with several dollars of credit remaining on them—were no longer being used.
“This helped us work out the most likely explanation: these cards had been lost, and the customers who owned them hadn’t purchased a replacement. So, we’re now looking at ways to refine our business model to make it easier for people to replace their cards easily if they need to.”
Seth Womble adds: “The dashboard system has also had an immediate impact, giving us a broader and more detailed overview of the situation at each of our projects, and allowing us to raise our standards even higher.
“The ability to measure water flow for each of our projects, for example, allows us to move the conversation forward in interesting ways. Is this community generating enough sales from water consumption to build up sufficient savings to maintain their infrastructure, for example? If not, how can we work with them to help them increase their savings? The insight helps us work with our communities to decide on the best course of action.”
Sandy Funk says: “Looking at the bigger picture, the ability to dive deeper into our data won’t just help the communities we serve—it will also support our scientific research into the broader problems of the global water crisis. For example, we will be able to monitor the effects of weather on water usage in different areas, and assess how well communities are equipped to deal with drought. We present every year at conferences like the UNC Chapel Hill Water Institute conference and Stockholm World Water Week, so research is a key part of our mission.”
George Greene IV concludes: “IBM Analytics helps us improve transparency and accountability—showing our donors how effectively we are stewarding their gifts, and what a difference we’re making to people’s lives. It also helps us demonstrate to the NGOs we work with, such as UNICEF and UNHCR, that our methods and technology really work, and that we are genuinely delivering on our promise of providing safe, sustainable water supplies for communities around the world.”
About Water Mission
Founded in 2001, Water Mission is a non-profit organization that specializes in environmental engineering solutions that help communities obtain sustainable access to clean, safe water. To date, Water Mission has served more than 3,000 communities, completed more than 1,000 solar pumping installations, and has more than 250 staff members working in over 50 countries around the world.
- Analysis / Modeling / Design & Construction
- IBM Analytics for Apache Spark (Managed Cloud Spark Service)
- IBM Cloud Databases and Compose
- IBM Cloud Managed Services
Take the next step
IBM Watson Data Platform offers a complete portfolio of data and analytics services providing unique and seamless product integrations to build apps faster and gain new insights easier with flexible deployment and pricing options. For more information about how IBM Watson Data Platform can help businesses solve tough big data problems rapidly and cost-effectively, please visit ibm.com/analytics/us/en/watson-data-platform/.