Analytics for Volunteer Emergency Response

Reducing emergency response times is of paramount importance, and traditional emergency services can be helped by volunteers that are dispatched via a smartphone app. Such community first responder (CFR) systems are already commonly used to send resuscitation-trained volunteers to cardiac arrest patients, and may in the near future also be used for other emergencies. In this unique project you work closely with CFR systems, ambulance providers, firefighters, red cross and others.

In this PhD project you will develop methodologies and algorithms to ensure CFR systems operate in the most efficient and effective way. This includes

• modeling the effectiveness of such systems in relation to the number of volunteers
• the spatial optimization of where to recruit volunteers
• developing alerting strategies: which volunteers should get an alert? Should we alert many immediately, or build in a time lag?
• How to best spend a training budget to improve the skills of some of the existing volunteers
• optimal volunteer task diversification: which volunteers do you instruct to travel directly to the incident and which ones need to pick up equipment first?

In the models you will develop, stochasticity plays an important role, as both volunteer locations and whether or not they will accept their task are uncertain. Techniques may include mathematical optimization, simulation and machine learning.

Faculty to contact: Dr. Pieter van den Berg

Volunteers, Emergency response, Operations Research, Machine Learning

Reducing emergency response times is of paramount importance, as there is a direct relation between response time and survival (Blackwell, 2002) or damage (Challands, 2010). Traditionally, emergency services consist of ambulances, firefighters and police, but these professionals can be helped by Community First Responders (CFRs): nearby, trained volunteers that are dispatched via a smartphone app. In severe cases, volunteers can cover those first crucial moments before the emergency services arrive; in milder cases they may even replace professionals and relieve work pressure on currently highly overloaded emergency services. The introduction of CFRs expands the chain of emergency response and requires rethinking of traditional emergency logistics.

There are already a few CFR systems that complement the Dutch emergency services: an app called HartslagNu dispatches volunteers to perform resuscitation on cardiac arrest victims. Burgernet supports the police and is typically used to retrieve missing persons. Finally, Ready2Help by Red Cross dispatches untrained volunteers for handyman services. In the Netherlands, more than 1% of the population is registered as a volunteer on at least one of these platforms.

We observe an emerging trend where parties indicate the ambition to increase their CFR usage or initiate a new CFR system. For example, Fire Department Amsterdam-Amstelland has included the ambition to have a network of 100,000 volunteers by 2030 (Legemaate 2022). Red Cross has, besides their untrained volunteer base at Ready2Help, a set of medically trained volunteers which they currently only deploy at events such as festivals and parades. They wish to explore the value of an app that would dispatch these same volunteers to non-life-threatening yet urgent incidents, such as helping up an elderly citizen that requires assistance after falling.

The SPRINTER project is a collaboration between three knowledge institutes and firefighters, ambulance services, a system that dispatches urgent help to elderly (Beep) as well as Red Cross, Burgernet and a UK-based system that dispatches highly trained volunteers (LIVES). As we have parties involved in different stages of implementation, there is a huge opportunity for cross-fertilization. Developed models will be made available in easy-to-use proof-of-concept software that allows for party-specific analysis.

SPRINTER proposes prescriptive analytics for optimal usage of CFR systems. We distinguish five work packages: WP1 concerns the management of the project and building a community of practice for CFR. WP2 analyzes the effectiveness of CFR systems by predicting the attainable response time and identifying volunteer tasks. WP3 focuses on the optimal use of budget for geospatial recruitment and training of volunteers. WP4 optimizes the real-time decisions related to alerting. Finally, in WP5 a proof of concept will be designed, implemented and tested.

There are currently no good methods available to support strategic and operational decision-making within these systems. This results in uninformed decisions, with risks of introducing a system that is not effective or does not live up to its potential in terms of impact. Worldwide, there is scarce literature on analytics for community first response; the few articles that exist mainly focus on simulation. We recognize an opportunity to advance the state of the art by developing new methods that go beyond the evaluative capabilities of simulation: SPRINTER proposes to develop mathematical models that proactively and quantitatively support the logistics of CFR systems.

The models developed in this project will be implemented in decision-support software that can be used by parties that have a CFR app or consider building one. The models will use innovative algorithms based on optimization and machine learning, made available via proof-of-concept software to 1) quantify the impact of volunteers, 2) decide where to recruit and train volunteers, and 3) decide which volunteer to alert when. The novelty of viewing these questions through an optimization lens results in ample opportunities for scientific publications; the practical value of SPRINTER is to advance the Netherlands to a point where parties are ready for a wide-spread use of CFRs.

Mathematical Modelling, Machine Learning, Simulation, Optimization

Required by you

• A master in operations research, operations management, business analytics, mathematics, industrial engineering, computer science or similar 

• Programming skills
• Good communication skills in English
• Proficiency in Dutch is an advantage because of our Dutch project partners

Required by ERIM

All application documents required by ERIM can be found here: https://www.erim.eur.nl/doctoral-programme/phd-in-management/admissions/application/

Below are the test scores required by ERIM:

• GMAT/GRE scores above the 85^th percentile, not older than 5 years
• IELTS/TOEFL test scores, not older than 2 years:

• IELTS: minimum overall band score 7.5; no band score lower than 6.5
• Internet-based TOEFL: minimum total score 100, no sub-score lower than 23
• Paper-based TOEFL: minimum score 600

The successful applicant is expected to start between December 1st and February 1st

Applications will be reviewed on a rolling basis until 30 September 2023.

We expect you to, together with the supervisory team, conduct research that is publishable in top academic journals. See https://www.erim.eur.nl/about-erim/erim-journals-list-ejl/ for a list of Primary Star (P*), and Primary (P) journals.

Besides the academic output, we expect the project to deliver proof-of-concept decision-support software that can be used by parties operating an CFR app.

This PhD position is part of the SPRINTER project, which is a collaboration between three knowledge institutes and firefighters, ambulance services, a system that dispatches urgent help to elderly (Beep) as well as Red Cross, Burgernet and a UK-based system that dispatches highly trained volunteers (LIVES). During the PhD project, you are expected to work closely with both the academic partners as well as the industry partners.  

The close collaboration with a consortium of industry partners in the emergency response sector ensures the societal relevance of the project. We expect you to work closely with these partners to make real-life practical impact. The envisioned supervisory team has a strong record on having practical impact with high quality research.

The Supply Chain Management group at RSM has a strong track record of publishing research in highly ranked management journals. In addition to undertaking research visits, PhD students are actively encouraged to participate in all major international conferences, and are supported in their travel to disseminate their research findings.

See the following two papers for an impression of the type of research conducted in this project.

van den Berg, Pieter L. and Henderson, Shane G and Jagtenberg, Caroline J. and Li, Hemeng, Modeling Emergency Medical Service Volunteer Response (April 11, 2021). Available at SSRN: https://ssrn.com/abstract=3825060 or http://dx.doi.org/10.2139/ssrn.3825060

Henderson, Shane G. and van den Berg, Pieter L and Jagtenberg, Caroline J. and Li, Hemeng. How should volunteers be dispatched to out-of-hospital cardiac arrest cases?. Queueing Syst 100, 437–439 (2022). https://doi.org/10.1007/s11134-022-09752-z

ERIM offers fully-funded and salaried PhD positions, which means that accepted PhD candidates become employees (promovendi) of Erasmus University Rotterdam. Salary and benefits are in accordance with the Collective Labour Agreement for Dutch Universities (CAO).

Erasmus University Rotterdam aspires to be an equitable and inclusive community. We nurture an open culture, where everyone is supported to fulfil their full potential. We see inclusivity of talent as the basis of our successes, and the diversity of perspectives and people as a highly valued outcome. EUR provides equal opportunities to all employees and applicants regardless of gender identity or expression, sexual orientation, religion, ethnicity, age, neurodiversity, functional impairment, citizenship, or any other aspect which makes them unique. We look forward to welcoming you to our community.

For questions regarding the PhD application and selection procedure, please check the Admissions or send us an e-mail via phdadmissions@erim.eur.nl.

ERIM is the research institute of management at Erasmus University Rotterdam, founded by Rotterdam School of Management, Erasmus University (RSM) and Erasmus School of Economics (ESE), both of which excel in their research and teaching programmes. The level and international character of their research and teaching programmes has put both schools at the forefront of academic excellence. They offer a diversified portfolio of accredited teaching and research programmes, ranging from pre-experience MSc to postgraduate Masters and executive development, and from fundamental research to applied research and business support, which are renowned throughout the world. RSM is one of the few schools worldwide that holds a triple accreditation from the AMBAs, EQUIS and the AASCB, and is positioned among the top European business schools. The scientific staff of both schools plays a major role in the schools’ interaction with the international business community.

ERIM distinguishes itself by the high-level and international character of its research and has repeatedly been ranked among top 3 research institutes in Europe. ERIM’s key aims are top-quality academic research with impact both on academia and managerial practice, and the advancement of international academic careers and high-level support for both research and doctoral education. For more information about ERIM please visit [link:https://www.erim.eur.nl].