On 29 and 30 October 2026, the inaugural ACM Conference on Digital Transformation, ACM DXConf 2026, took place at the Michigan League in Ann Arbor, Michigan. Hosted by the ACM Emerging Interest Group on Digital Transformation (EIGDX), the new conference sets out to establish digital transformation as a discipline of its own, bringing computing researchers together with people working on water, energy, agriculture, manufacturing, transport and digital infrastructure.

RESCUE coordinator Anders Lindgren (RISE Research Institutes of Sweden) was Program Co-Chair of the conference together with Andreas Blenk (Siemens), with Muhammad Shahbaz (University of Michigan) as General Chair. Two papers from joint work between RISE and Sensative were presented, both closely connected to the RESCUE goal of keeping water and urban services running.

Two days of discussion across disciplines
Over two days, 14 accepted papers were presented in four technical sessions, alongside three keynotes and a panel on the background and road ahead for the new ACM community on digital transformation. The sessions covered public infrastructure and policy, power and grid-integrated computing, water and community utilities, and operational intelligence at the edge and in industrial systems. Anders chaired the session on power, energy and grid-integrated computing, took part in the panel and gave the closing remarks.

The panel turned into an open community discussion on where the field should go next: which abstractions, methods and benchmarks carry across grids, water, manufacturing, transport and supply chains; how to bring industry and academia closer through shared data, testbeds and experience papers; and whether sector-focused workshops, for example on water, could help grow communities around specific infrastructures.

The two days were marked by very good discussions and presentations, a high level of interaction and an interesting mix of disciplines. In his closing remarks, Anders pointed to three threads that ran through the programme: AI only scales when it is tied to real data, engineering context and accountable execution; computing and power systems increasingly need to share models, interfaces and responsibility; and operators adopt what they can understand and verify, which puts interpretability, privacy and long-term field experience at the centre.

Privacy-preserving leak detection across water utilities
In the Water and Community Utilities session, Anders presented "Privacy-Preserving Leak Detection Across Water Utilities via Secure Multi-Party Statistics" (Anders Lindgren and David Eklund, RISE, and Sara Moricz, Sensative). Water utilities lose large volumes of water to leaks, and detection improves when utilities can compare notes. But sharing raw data is not an option: beyond being commercially sensitive, it would reveal the layout and weak points of critical infrastructure.
In the approach presented, each utility turns whatever measurements it has (pressure, flow or minimum night flow) into a standardised leak indicator, and the utilities then compute shared statistics using secure multi-party computation, so that only the combined result is revealed. The group learns the average and spread of the indicators and the number of alarms, and each utility can privately see how it ranks against the others without seeing anyone else’s values. The work uses a reproducible simulation environment with water networks derived from real street maps. On twelve district-scale networks, the secure results matched an ordinary, non-private computation almost exactly, and the paper also takes first steps towards detecting tampering and rejecting manipulated inputs. In short: the representation protects the network layout, and the cryptography protects the data.
The paper is available from the ACM Digital Library: https://doi.org/10.1145/3847238.3849047
Real-time indoor localisation from Wi-Fi signals
Justus Kluge (KTH Royal Institute of Technology) presented "Real-Time Indoor Localisation using Passively Captured Wi-Fi Probe Requests" (Justus Kluge, Benedek Boldizsár, Fredrik Westman, Anders Lindgren and David Eklund), a collaboration between KTH, Sensative and RISE. Knowing how buildings and public spaces are used matters for heating, ventilation and security, but cameras are intrusive and heat sensors cover little area. The paper takes a middle road: low-cost ESP32 sensors passively pick up the Wi-Fi probe requests that phones send anyway, and an edge server turns them into position estimates in real time using a propagation model of the building’s rooms and walls that needs only a short calibration. In tests in a 470 m² and a 200 m² space, the median error was about 2 and 3 metres, and devices were placed in the right room about 87% of the time. Processing stays at the edge, so personal data stays local, and room-level occupancy can also help spot unexpected activity, such as someone in an office that should be empty at night.
The paper is available from the ACM Digital Library: https://doi.org/10.1145/3847238.3849046
Looking ahead
For RESCUE, DXConf was a good place to meet researchers and practitioners working on the same questions from different angles: how to make digital infrastructure for water, energy and cities resilient, trustworthy and useful in practice. The next DXConf will be held on 11 to 12 October 2027 in Eindhoven, the Netherlands, hosted by Eindhoven University of Technology, with a submission deadline of 7 April 2027. More about the community and how to join is available at https://sigdx.org.
This project has received funding from the European Union’s Horizon Europe research and innovation programme under grant agreement No 101225910.