Current projects

Wings For Aid, a community powered by experts, is developing a Remotely Piloted Aircraft System that delivers humanitarian goods to people isolated by natural disasters and man-made crises — safely and with pinpoint accuracy.
The Hightech Business and Entrepreneurship department of University of Twente supports the Wings For Aid program with the modelling, optimization and analysis of the logistics processes and drone operations. This includes the development of a generic simulation model for UAV-aided humanitarian logistics. The model contains a library of disasters and their essential attributes, allowing experimentation with, e.g., various supply chain configurations and drone operations. Point of contact for Wings For Aid is dr.ir. Martijn Mes.
Presentation from Robert van Steenbergen at the Winter Simulation Conference 2020

The project "Operations Research and AI for Airline Operations" is a collaboration between UT and KLM Royal Dutch Airlines as part of the Kickstart AI initiative, involving a jointly funded assistant professorship (Dennis Prak) and two jointly funded PhD candidates (Antonio Montaruli and Mahekha Dahanayaka). This project includes various research and education activities, notably:
the PhD projects of Antonio Montaruli and Mahekha Dahanayaka, embedded at the section IEMS at UT and the Data Science platforms Integrated Planning & Control and Ground at KLM. Topics include strategic and tactical airline network planning, airline network revenue management, and workforce planning under operational uncertainties. The aim is to obtain practicable decision support for complex airline network problems, using novel prediction techniques, stochastic and robust optimization methodologies.the minor module Aerospace Management & Operations, coordinated by Dennis Prak, which offers a broad introduction into the aviation industry, airline and airport management, open to all UT BSc students and exchange students.various IEM BSc and MSc thesis topics on related subjects, such as robust maintenance slot planning, missed connection risk prediction, mitigating gate delays due to hand luggage overflow, and estimating durations and volatilities of platform handling tasks.
This project is funded by KLM Royal Dutch Airlines.

The increasing impact of climate change – such as drought, flooding and damage to infrastructure – is putting pressure on the reliability of logistics networks. Multimodal corridors, which depend on water, rail and road, are particularly vulnerable to disruption. If these routes are temporarily unusable, there is a risk of a reverse modal shift: goods will once again be transported en masse by road, with negative consequences for the environment and the efficiency of the network. Changes to physical infrastructure are often slow and costly. It is therefore crucial to develop smarter, resilient logistics systems that can respond flexibly to changes.
The project ‘Transitioning towards resilient multimodal corridors using digital twinning’ (TRIUMPH) is working on a digital environment – a so-called digital twin – that can be used to map, simulate and improve logistics corridors. By using data (such as weather information, logistics planning and real-time sensor data), companies and governments can better anticipate risks. The project focuses on designing practical tools (the “resilience toolbox”) and developing a digital model of a corridor, which partners can use to jointly determine which mode of transport is most suitable at any given time. This creates a climate-proof and flexible system that is better able to cope with unexpected situations.
The TRIUMPH project delivers two concrete innovations:
A resilience toolbox: a collection of practical strategies and methods that companies and governments can use to make their logistics networks more resilient to disruptions.A digital twin of multimodal corridors: a platform that enables real-time decisions to be made about the best transport combinations, depending on the situation.
These solutions are being tested in practice, including in the Twente corridor. In addition, the results are being shared more widely with the sector through workshops, publications and training courses. By making the digital tools more widely applicable, other corridors in the Netherlands and Europe can also benefit.

Problem context
The increasing pressure on the Dutch healthcare sector, caused by the ageing of the population, is leading to a growing gap between demand for care and the available workforce. Policy measures encourage independent living at home, which aligns with clients’ preferences but simultaneously creates new logistical and organisational challenges. The transition from care institutions to home care requires small-scale deliveries of care resources and services to the home, resulting in more transport movements, higher costs and a heavier burden on care providers. There is a need for a system that balances both clients’ preferences and the workload of care providers.

Approach
Within the Ontzorgen project, a regional care system is developed in which care, goods and services are coordinated in an integrated manner. The approach involves mapping client preferences, experiences of care providers and opportunities for task redistribution. Focus lies especially on staff, who take over non-care-related tasks such as the delivery of medical aids. In addition, research is being conducted into how deliveries can be consolidated and routes optimised to reduce transport movements and CO₂ emissions.
Expected outcomes
The project will result in the development of a regional control centre that supports real-time, data-driven decision-making within the healthcare ecosystem. This centre will facilitate the joint management of healthcare, goods and services supply chains, taking into account supply chain constraints and the preferences of those involved.
Concrete expected outcomes are:
Reduced workload for care providers through smart task allocationBetter alignment of care moments with client preferencesReduction in transport movements and emissionsA digital control centre for data-driven decision-making
The solutions will be tested in practice-oriented pilots with project partners, in preparation for wider implementation.
The consortium
The consortium comprises healthcare organisations, suppliers, research institutions and a network organisation. This combination brings together practical experience, data expertise and innovation capacity. Through this multidisciplinary collaboration, a bridge is built between research and implementation, which enhances the practical applicability and acceptance of the solutions developed.
Project partners: Care Invest Group, Carel Lurvink, Draaijer Group, Essity, Huuskes, Lucrum, Lyreco, Mediq, Pinkroccade, Renewi, TNO, Zorgcentrum de Posten, Zorgfederatie Oldenzaal, Zorgorganisatie Norschoten, Universiteit Twente, Windesheim
This project is funded by TKI Dinalog.

Existing heavily industrialized areas are currently incapable of adopting large-scale industrial symbioses in terms of shared technology/ infrastructure use, waste integration, energy and material utilization, as well as expanding through surrounding ecosystems in an inclusive manner for the society despite many of the EU's key strategic priorities in sustainable regional development. Within this context, IS2H4C proposes an ambitious and efficient innovation and action workplan to develop several solutions for the development of Hubs for Circularity (H4C) in diverse industrial areas of process industry surrounded by rural and/or urban settings in the Netherlands, Germany, Spain, and Turkey. The workplan is shaped by development and deployment of most innovative sustainable technologies and infrastructure integration in four demo hubs and is supported by ground-breaking research on societal, governmental, and business innovation for H4C. IS2H4C scales up industrial areas to H4C via implementing systemic change and integrating the surrounding ecosystems to industrial areas. The project implements a digital collaboration platform to manage the resource, infrastructure and information-sharing within H4C via embedding of decision-support modules in the platform. IS2H4C has the ambition of reducing energy use, waste emissions, and carbon emissions by at least 10%, 20%, and 30% respectively. IS2H4C will contribute to pave the path towards the development of H4C based on the circularity and resilience requirements of existing and future industrial zones and their surrounding ecosystems by prioritizing resource efficiency, maximizing use of renewable energy, prevention of waste, and promoting industrial/urban/rural symbiosis via reuse and recycling of unavoidable solid, liquid, and gas waste streams. All in all, IS2H4C’s biggest ambition is to promote H4C as Europe’s future sustainable regional development models.

In recent years, the energy sector has undergone changes that have a high impact on the dynamics in power markets. One of the changes has been the shift towards energy production from intermittent sources, such as wind and solar. As a consequence of this shift, the amounts of energy that are traded at the short-term markets throughout a day are uncertain, as they depend on hardly predictable weather conditions.
This uncertainty increases the volatility of short-term energy prices, and thus makes it much more challenging to make economically viable energy trading decisions. One way to respond to this challenge is to leverage assets such as grid-level battery storage, and electrolyzers to have more flexibility when making trading decisions. The challenge then is how to optimally leverage such an asset to make viable trading decisions under high price volatility. This research project focuses on designing, developing, and evaluating self-learning energy trading algorithms that are able to cope with these challenges. By leveraging real-time data, developed algorithms continuously adapt to market dynamics and respond to changing market signals with economically viable trading decisions.
This project is funded by EWE Trading.

The project’s objective is to develop and evaluate healthcare software products and algorithms that improve the streamlining of patient flows and capacities in the program area, ensuring access to and efficiency of healthcare. We focus on emergency services, hospitals, and care facilities and their collaboration across the border. Due to the geographical proximity in the border region, the duplication of, e.g., ambulances and (intensive care) beds is not absolutely necessary. Rather, the joint use of capacities across the border offers the opportunity to provide patients with the care they need and to reduce overcapacity. This is a complex planning task that requires a substantial amount of coordination. We support this planning task. We use methods from the fields of IT, artificial intelligence, and analytics to ultimately manage patient flows between healthcare facilities and capacities within the region intelligently using software and algorithms. The algorithms enable data-driven decisions based on demand forecasts. Besides regular demand, we also consider peak demand, e.g., pandemics, as cross-border collaboration is even more important due to the high demand. We develop the software based on feedback from our partners in the healthcare sector in order to remove barriers for practical implementation. We simplify the use of the products through a new training concept. The cross-border perspective on patient flows and capacities offers great potential to improve efficiency, quality, and health outcomes in the program area. The cross-border consortium comprises companies, universities, and partners from the healthcare sector and combines the necessary expertise to address requirements from Germany and the Netherland.

In recent years, online shopping has for many consumers become the standard way of purchasing products. Yet, the number of online orders continues to increase strongly as more and more consumers switch to online shopping of fast-moving consumer goods (FMCG), such as packaged foods, beverages, and non-durable household goods. As a consequence of this trend, FMCG delivery services face the challenge of having to process large numbers of orders at a high pace to guarantee fast delivery.
The key to meeting this challenge is to maximize the efficiency of last-mile delivery operations. To this end, delivery services need methods that are able to derive optimal decisions about vehicle routing and order picking in the course of each business day. Developing such methods is challenging, as optimal decision making requires taking into account the uncertainty about upcoming customer orders of the remaining day.
This research project focuses on designing, developing, and evaluating self-learning optimization algorithms that are able to cope with this challenge. By leveraging real-time data, developed algorithms continuously adapt to customer behavior and respond to changing market signals with economically efficient routing and picking decisions.
This project is funded by Flaschenpost SE.

Patients with chronic conditions are often scheduled for routine follow-up appointments at fixed intervals, even when these visits are not always medically necessary. Research shows that over 18% of such consultations can be avoided (Klaas et al., 2024). The WACHT-NIET project therefore focuses on a new, data-driven approach to scheduling, where follow-up appointments are based on medical necessity as well as patient preferences.

In collaboration with general practitioners, patients, and healthcare partners, WACHT-NIET develops an integrated invitation and scheduling system for patients with chronic kidney disease (CKD). Based on laboratory results and patient questionnaires, the system determines whether a follow-up consultation is needed or desired by the patient. Appointments are then only scheduled when necessary, with the goal of ensuring timely access to care (target: 98% of patients seen within one week).

The project combines data analysis, predictive modeling, and optimization algorithms to improve both clinical decision-making and operational planning. It also addresses the appropriate distribution of care between primary and secondary care, and explores risk-based follow-up instead of fixed scheduling intervals. The approach is being tested in practice at the Isala hospital and participating general practices. Key outcomes include access times, number of consultations, workforce efficiency, and patient satisfaction.

WACHT-NIET contributes to more sustainable healthcare by reducing unnecessary appointments, improving the use of available capacity, and ensuring that patients receive the right care at the right time.
Project partners: University of Twente, Isala, Medrie
Funding: ZonMw MEDisch specialistische Zorg en Onderzoek (MedZO) Lijn 1: Passende en doelmatige zorg SA 2024

The logistics sector in the Netherlands is a vital economic pillar, employing over 673,000 people and contributing €65 billion annually. However, the sector faces pressing challenges, such as reducing greenhouse gas emissions, ensuring supply chain resilience amidst disruptions, and overcoming infrastructure and workforce shortages. With freight transport projected to grow by 20% by 2030, these challenges require a shift from isolated logistics operations to collaborative, connected logistics networks. Upcoming policy measures, including kilometre chargers, CO2 caps, the Emissions Trading Scheme, and the Corporate Sustainability Reporting Directive, add urgency to this transition.
A promising framework for addressing these challenges is the Physical Internet (PI), which envisions a transformative shift in logistics systems. The PI concept aims to “do more with less” by enabling the sharing of assets within the freight and transport industries. This involves transitioning from the isolated scheduling of proprietary assets to the collaborative scheduling of shared resources in open, connected logistics networks.
This project is part of TNO’s Early Research Program on Future-Proof Smart Logistics. It aims to contribute to the realisation of the PI concept by developing advanced machine learning-based decentralised decision-making algorithms. These algorithms will enable logistics companies to collaborate effectively and optimise operational scheduling across multi-actor systems, ensuring sustainability and efficiency at both company and system levels.
This project is funded by TNO.

The project Predictive Workload-Aware Feature-Based Planning for Resilient High-Mix Low-Volume Manufacturing (PREDICTIVE PLAN), is a strategic initiative within the High Tech Systems and Materials (HTSM) program, specifically focused on Systems Engineering for High Tech Systems.
The primary objective of PREDICTIVE PLAN is to develop and validate a predictive, feature-based, and human-centric planning framework designed to manage variability early and throughout the production process. It aims to provide explainable, data-driven digital assistance to planners and engineers by combining artificial intelligence, simulation, and human expertise to enhance the sustainability, resilience, and efficiency of Dutch high-tech manufacturing.
The project addresses the critical challenges faced by High-Mix, Low-Volume (HMLV) manufacturers like Thales NL. Current planning systems (ERP/APS) are designed for standardized products and fail to account for high Variability due to unique products and complex routings lead to unpredictable bottlenecks and resource conflicts, human capital risk, planning heavily relies on the "best-guess" estimations of experienced planners who are nearing retirement, making the function vulnerable and unsustainable
Predictive Plan follows a 4-Step modular research cycle to move from proof-of-concept to industrial implementation. Including Digital Twin (Module 1): Building a discrete-event simulation to model demand, supply, and operational variability, AI Surrogate Models (Module 2): Developing machine learning models for near-instant predictions of workload, waiting times, and throughput, Hybrid AI Planning (Module 3): Integrating surrogate predictions with reinforcement learning and heuristics to optimize job release and material planning, Human-in-the-Loop (Module 4): Implementing a framework that uses Reinforcement Learning from Human Feedback (RLHF) to capture tacit knowledge and ensure planners trust and adopt the system.
The project targets reaching a Technology Readiness Level of 7 (TRL7) with a validated prototype operating in "shadow mode" at Thales, Key performance targets include: 40–50% reduction in waiting times and work-in-progress (WIP), 10–15% reduction in material waste and energy consumption, increased on-time delivery and faster, more consistent planning decisions and digital Intelligence for human workers: Capturing the expertise of retiring planners to support younger staff with explainable AI.
This project is funded by TKI HTSM.