In The Netherlands “the land of water” the ecological quality of ponds, ditches, wetlands and lakes is severely degraded due to escalating and interacting anthropogenic pressures including pollutants and climate change. SEFAP unites leading Dutch freshwater experimentalists, infrastructures and data scientists to provide a step forward in collaborative science and inland water ecology. By conducting experiments in SMART-enabled replicated mini-lake ecosystems, SEFAP will enable the future of our waters to be experimentally created and tested. In combination, the technical innovation and community-building of Dutch aquatic experimentalists will strengthen the ability to predict and mitigate undesirable futures in aquatic ecosystems.

Power4FitFoot focuses on an ecosystem to support data-driven personalized and persuasive monitoring & coaching for cardiovascular patients with diabetic foot. The result will be an early warning system (EWS) on risk detection of deterioration (healthcare providers), with feedback and coaching (patients and informal caregivers). This project shows that the combination of big data technologies and real-time sensor data analytics is the key ingredient to tailor and personalize self-management programs. Power4FitFoot follows a participatory multidisciplinary development approach: researchers (medicine, computer and behavioral sciences) work together with patients, caregivers and industrial companies to develop personalized self-management products and services to prevent Diabetic Foot Ulcers (DFU) or amputations. We outline 4 complementary studies to accomplish our goals. Study 1 focuses on the identification of the major risk factors using existing retrospective data sets. Outcomes will be transformed into dependent variables we aim to predict. Study 2 entails the co-creation of a smart monitoring ecosystem (platform & body sensors) to enable personalized feedback via data analyses. Data sets from various sources (medical, lifestyle, geo-spatial) and with high variety will be collected real-time to further develop the prognostic models. Study 3 focuses on real-time coaching dealing with high risk contexts using the insights of the big data sets from both previous studies. Prognostic models will facilitate a dynamic and context aware heuristics for a self-management support system. Study 4 is an overarching study, to guarantee the infrastructure for data-driven monitoring & coaching and to develop a sustainable data management plan.

Biodiversity loss is a critical global challenge. Understanding and monitoring biodiversity are essential for policymaking and ecosystem resilience. Current monitoring systems are restricted, and collecting and analyzing data is time-consuming, costly, and challenging to scale. Our solution frees up time and enables monitoring more extensive areas with higher resolution and longer durations. We achieve this by automating routine tasks using embedded AI on the sensor device. In this project, we are validating our target market and the added value of embedded AI through an intelligent sensor for avian nestboxes that autonomously detects and counts the prey captured by birds.

The bicycle industry is undergoing a radical transformation - the trend towards electrification and digitisation in combination with shared mobility concepts and heightened customer expectations requires manufacturers to develop novel approaches for product innovation, manufacturing and service delivery. The influx of new technologies and materials is rapidly changing the appearance, function and role of bicycles and E-bikes and custom bikes now represent a significant part of the bicycle market. Effective incorporation of bikes into the future connected and smart transport networks require technologies and mechanisms to allow monitoring the bike, understanding the cyclist and the context. The project will advance the industry in improving lead-time and effectiveness of their product innovation process, as well as the functionality, reliability and quality of their products and services.

To reduce the down time and associated costs in a production facility, the predictability of failures is very important. This can be achieved when firstly the relevant parameters on operational conditions are monitored and secondly that data is properly processed to obtain accurate estimates of time to failure. In this project, the first challenge will be addressed by developing advanced wireless sensor networks, that enable the collection of the right data in a flexible way. The second challenge will be addressed by the development of physical failure models for the most critical components in the system. By feeding the models with the monitored variation in operational settings, the time to failure can be predicted and appropriate maintenance tasks can be scheduled. For less critical components a more data-driven approach will be followed, resulting in a decision support tool enabling optimization of the maintenance process and maximizing plant uptime.