Let’s Get Moving and Write Software: An Open Source Project for EPANET
Publication: Journal of Water Resources Planning and Management
Volume 144, Issue 4
Introduction
Water infrastructure simulation models, such as EPANET and SWMM, have played an important role in the development and dissemination of simulation technologies for water distribution, stormwater and sanitary sewer systems. These software packages have performed varying functions for different portions of the water infrastructure community. Municipalities, water utilities, and consultants use these packages to ensure adequate water services and to protect public health. Commercial vendors have integrated and enhanced the core software codes within their products, extended their capabilities, and enhanced the end-user experience. Academicians and researchers have modified and extended the source code algorithms to improve the core modeling components and develop new tools and applications. A primary reason for the success of these packages (EPANET and SWMM have joint average downloads of approximately 100,000 per year) is that the software and source code have been freely available and evolved to satisfy a range of needs, from those of a typical end-user (graphical user interface) to those of a researcher (robust algorithms and interfaces).
Although these packages remain successful and the global water infrastructure modeling community has been productive, the overall community associated with these free, open-source software packages remains fragmented with respect to development. For example, whereas development of the EPANET Multi-Species eXtension (EPANET-MSX; Shang et al. 2007) has been included with the official release of EPANET, previous developments associated with object-oriented versions of the hydraulic/water quality network solvers (e.g., van Zyl et al. 2003; Guidolin et al. 2013), pressure-driven solvers (e.g., Giustolisi and Laucelli 2011; Mahmoud et al. 2017), or transient hydraulic simulators (e.g., Ramalingam et al. 2009; Zecchin et al. 2010) have largely remained the domain of their developers. Although the overall volume of research output associated with EPANET is impressive, there has been little formal organization around highly specific research themes, and little effort expended on producing high-quality, end-user software products associated with EPANET.
As one example, a literature search about network model calibration yields hundreds of articles on techniques and applications for water distribution models, dating from 1978. The American Water Works Association is working on its fourth update to the M32 manual on water distribution system modeling and related calibration issues (AWWA 2012). Yet the most recent periodic survey of utility stakeholders shows that the routine aspects of model calibration ranked as the most technically challenging modeling task (AWWA 2014). Although calibration algorithms have been implemented in commercial products, there is a clear need for additional solutions—some of which may be found among the many lines of computer code written over the decades in service of model calibration research. Clearly, those modifying the EPANET source code base for research purposes have failed to coordinate the significant code development efforts into jointly developing, improving, and maintaining high quality calibration software that the practice communities need. This failure of coordination is not a minor problem and has hampered the ability to stand on each other’s shoulders, and move the field of water infrastructure modeling forward at the pace demanded by practical needs.
Where Did the Developer Go?
During the last 20+ years, the EPANET community (involving many people, from research and commercial sectors) has primarily left the task of converting infrastructure modeling research into a productive, free, and open-source software product to one person—Dr. Lew Rossman at the USEPA. This is at once a testimony to Dr. Rossman’s incredible productivity and skills and, frankly, to a lack of active community engagement and interest in EPANET software development. Although this arrangement has been comfortable, it no longer works. There has not been a major update to the flagship EPANET software since the official release of EPANET version 2.0 in 2000, as efforts continue to pour into the production of prototype code and research volumes—to be locked there forever, without impacting those that utilize EPANET. As a community focused on free and open source EPANET development, we can, and must, do better—the global community can no longer afford to spend scarce resources producing models and software that few people will use.
Call to Keyboards
In order to ensure the long-term sustainability of free, open-source water infrastructure simulation models, there is a need to engage and organize a global community around the research and development of models and of high quality software, for the analysis, design, and operation of complex water infrastructure systems. It can be done—the groundwater research and modeling communities have, for example, collaborated productively for 30 years on the development of novel processes and packages for groundwater flow and solute transport, with the user end-point being the industry-standard MODFLOW package (Harbaugh 2005). The latest version of MODFLOW includes four fundamental processes for groundwater flow, sensitivity analysis, field observation, and model calibration. More than 100 separate packages are available that codify the best hydrologic science, covering everything from aquifer subsidence, to large-scale multigrid solvers, to groundwater management. Such success is achieved through a focused organization, and software design decisions that produced a robust framework that fosters the integration of additional capabilities.
The same success achieved by the USGS for groundwater modeling can be achieved by a new community focused on water distribution system modeling. A modern approach to open source software development puts emphasis on community engagement first, through a formal software design process. That process fosters a public vetting of technologies, enhancements, and features, and leads to a software architecture that can serve as a robust integration framework—one that clears a path between novel infrastructure research and end user software tools, leading to multiple practical outcomes for the water infrastructure modeling community, and the public that depends on these infrastructure systems.
The time has come for the EPANET community to engage in the development of software that will benefit the water distribution system community now and in the future. To that end, members of the community—many of whom are involved with the Water Distribution Systems Analysis Committee (ASCE) and related symposia and/or the Computing and Control of the Water Industry Conference series—have already established an Open Source EPANET Initiative for the continued development of EPANET. As part of this initiative, current efforts are underway that already include a new release (EPANET version 2.1) and continuing development (EPANET version 2.2) of the EPANET Toolkit and preliminary assessment and discussions of EPANET version 3.0, which will become the first major—community developed—release of the popular water distribution modeling software package.
Whereas the notion of community development often sparks images of individuals programming away on a project out of a labor of love (which is true to some extent), community development needs to, and must, be much broader. To be successful, community development requires the contributions from a wide range of individuals with varying interests in development and/or end-use. Clearly, the contributions of those interested in software development, enhancing the base code, and contributing their own developments are critical for ensuring that the underlying architecture and new techniques are incorporated into the EPANET software product. However, there are many other opportunities to contribute that are just as valuable and can be achieved by the broader community. These additional opportunities include items such as beta testing new versions/features and reporting bugs, improving documentation and providing illustrative examples, and suggesting new features that could be developed and incorporated into the EPANET software. The latter contribution is particularly important to receive feedback from both end-users and code developers about potential new features to help shape the future direction of EPANET development. Thus, there is a potential role for everyone, which includes the software developers, researchers and students, software vendors, government and utility personnel, and consultants to contribute in their own way to ensure successful community development of the EPANET software.
If this argument speaks to you—there is now an opportunity for everyone to become engaged as part of the open source water infrastructure modeling community that will allow your efforts and opinions to be shared with a wider community. To become a part of that community and become instrumental to its success go to Open Water Analytics (2018)—sign up and participate in the discussions; view the water infrastructure software projects hosted on GitHub; watch a project that’s interesting to you; browse the list of current issues and code commits; and contribute your own code. Be a part of the software development efforts that will define the solutions for our industry. Be a part of our community’s future.
Disclaimer
The U.S. Environmental Protection Agency collaborated in the work described here. It has been subjected to the Agency’s review and has been approved for publication. Note that approval does not signify that the contents necessarily reflect the views of the Agency. Mention of trade names, products, or services does not convey official EPA approval, endorsement, or recommendation.
References
AWWA (American Water Works Association). (2012). M32 computer modeling of water distribution systems, 3rd Ed., Denver, 238.
AWWA Water Distribution Model Survey Subcommittee. (2014). “Committee report: Trends in water distribution systems modeling.” J. AWWA, 106(10), 51–59.
EPANET-MSX version 1.1 [Computer software]. U.S. Environmental Protection Agency, Washington, DC.
EPANET [Computer software]. U.S. Environmental Protection Agency, Washington, DC.
Giustolisi, O., and Laucelli, D. (2011). “Water distribution network pressure-driven analysis using the enhanced global gradient algorithm.” J. Water Resour. Plann. Manage., 498–510.
Guidolin, M., Kapelan, Z., and Savic, D. A. (2013). “Using high performance techniques to accelerate demand driven hydraulic solvers.” J. Hydroinf., 15(1), 38–54.
Harbaugh, A. W. (2005). MODFLOW-2005, the U.S. Geological Survey modular ground-water model: The ground-water flow process, U.S. Geological Survey, Reston, VA.
Mahmoud, H., Savic, D., and Kapelan, Z. (2017). “New pressure driven approach for modelling of water distribution networks.” J. Water Resour. Plann. Manage., 626–635.
Open Water Analytics. (2018). “Open source EPANET project.” ⟨http://wateranalytics.org⟩ (Jan. 16, 2018).
Ramalingam, D., Lingireddy, S., and Wood, D. J. (2009). “Using the WCM for transient modeling of water distribution networks.” J. Am. Water Works Assoc., 101(2), 75–89.
Shang, F., Uber, J. G., and Rossman, L. A. (2007). EPANET multi-species extension user’s manual, U.S. Environmental Protection Agency, Cincinnati.
SWMM version 5.1.012 [Computer software]. U.S. Environmental Protection Agency, Washington, DC.
Van Zyl, J. E., Borthwick, J., and Hardy, A. (2003). “OOTEN: An object-oriented programmers toolkit for EPANET.” Proc., Computing and Control in the Water Industry, London.
Zecchin, A. C., Lambert, M. F., Simpson, A. R., and White, L. B. (2010). “Frequency-domain modeling of transients in pipe networks with compound nodes using a Laplace-domain admittance matrix.” J. Hydraul. Eng., 739–755.
Information & Authors
Information
Published In
Journal of Water Resources Planning and Management
Volume 144 • Issue 4 • April 2018
Copyright
©2018 American Society of Civil Engineers.
History
Received: Sep 19, 2017
Accepted: Oct 6, 2017
Published online: Feb 2, 2018
Published in print: Apr 1, 2018
Discussion open until: Jul 2, 2018
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.