Skip to content
Dr Daniel Gartner

Dr Daniel Gartner

Senior Lecturer

School of Mathematics

Email:
gartnerd@cardiff.ac.uk
Telephone:
+44 (0)29 2087 0850
Location:
M/1.31, 21-23 Senghennydd Road, Cathays, Cardiff, CF24 4AG
Available for postgraduate supervision

My research interests are in mathematical programming, optimization and machine learning applied to the service industry, especially healthcare operations.

An author of more than a dozen of peer-reviewed journal articles and conference proceedings, my research was awarded with several prizes at international conferences such as INFORMS healthcare, INFORMS annual meeting, ORAHS and the OR Society's annual meeting and simulation workshop.

I am a Fellow of the Higher Education Academy and an Area Editor of two international journals on Healthcare OR.

Education

  • Fellow of the Higher Education Academy (Cardiff University)
  • Master in Medical Informatics (University of Heidelberg)
  • Master in Computer Science (Université Lyon I)
  • PhD in Operations Management (Technische Universität München)

Previous position

Post Doc in Healthcare Information Systems (Carnegie Mellon University)

2019

2018

2017

2016

2015

2014

I teach the following modules:

  • MA0261 – Operations Research
  • MAT003 – Communicating and Research Skills
  • MAT006 – Supply Chain Modelling

I'm supervising the following PhD students:

  • Emma Aspland (together with Prof. Paul Harper on a KESS2 funded scholarship)
  • Sarie Brice (with Prof. Paul Harper on an EPSRC funded scholarship)
  • Mark Tuson (with Prof. Paul Harper)
  • Emily Williams (together with Prof. Paul Harper on a KESS2 funded scholarship)
  • John Threlfall (together with Prof. Paul Harper on a KESS2 funded scholarship)

Research Interests and Overview

During my PhD, I applied scheduling methods to a health care operations management problem while taking into account uncertainty in patients’ lengths of stay and resource availability. However, to apply the models and methods in practice, many parameters have to be determined prior to the schedule calculation. Tackling this challenging problem was one of the tasks I carried out a post-doctoral fellowship at Carnegie Mellon University which is leading in the field of Machine Learning and Healthcare Information Systems. I’m still highly interested to connect these methods and here in Cardiff, I’m working with some collaborators in this area of research.

My overall research objective is to apply quantitative methods to a variety of problems on the strategic, tactical and operational decision levels of healthcare operations. I’m actively searching for collaborations writing grant proposal within the boundaries of the Engineering and Physical Sciences Research Council (which is similar to the Deutsche Forschungsgemeinschaft in Germany). I also strive for collaborations with local research authorities and innovation funds provided by, e.g. Welsh Government and the Health Boards in Wales (especially Aneurin Bevan).

Based on collaborations with Cornell University, Carnegie Mellon, UCLA but also Karlsruhe Institut of Technologie, I published papers in INFORMS journals, EJOR, Annals of Operations Research but also journals on the intersection between Health Care, Mathematics and Management (e.g. IMA Journal on Management Mathematics or Health Care Management Science).

Here is a short summary of current and past research projects focusing on healthcare:

Hospital-wide Patient Scheduling

The Job Shop Scheuling Problem is a combinatorial Optimization Problem. The goal is to schedule jobs on scarce resources, such that completion times or delays of jobs targets are minimized. Job Shop Scheduling is a classic problem in the manufacturing and service industries. An application is, for example, the machine layout planning. An extension of this problem is the resource-constraint project scheduling problem which considers order relationships between jobs.

In various projects I have developed Job Shop Scheduling models and their extensions for the efficient acquisition of healthcare services. This can also be applied to other industries, in particular steel production or flexible delivery to customers by parcel service providers.

  • „Scheduling the hospital-wide flow of elective patients.” European Journal of Operational Research, 223(3):689–699, 2014. (together with Prof. Dr. Rainer Kolisch)
  • “Handbook of Research on Healthcare Administration and Management, chapter: Mathematical Programming and Heuristics for Patient Scheduling in Hospitals: A Survey, Seite 627–645. IGI Global, 2017.” (together with Rema Padman, PhD)
  • “Exact and heuristic methods for the hospital-wide therapy scheduling and routing problem.” IIE Transactions on Healthcare Systems Engineering, 2018. (together with Prof. Dr. Rainer Kolisch and Dr. Markus Frey)

In another study we conducted with physicians at the University of Munich, we were able to show that Mathematical Programming can help to optimize the mix of reusable and disposable endoscopy equipment in hospitals. This problem can be classified at a strategic level.

  • “Cost-efficient employment of reusable and single-use systems through mathematical programming.” Anesthesia & Analgesia, 2017. (together with Dr. Günther Edenharter and Dr. Dominik Pförringer).

Machine Learning for the Strategic and Tactical Planning in Healthcare

In the provision of service processes, many parameters are subject to stochastic fluctuations. Strategic planning seeks to predict and aggregate customer flows with high accuracy. This problem is known from the area "Sequential Pattern Mining", which I solved in a work with a doctoral student from Karlsruhe at the chair of Prof. Stefan Nickel. Based on a layout planning problem, the goal was to predict patient flows as accurately as possible and to optimize a hospital layout.

  • „Improving Hospital Layout Planning Through Clinical Pathway Mining.” Annals of Operations Research, 2018, Volume 263, Issue 1-2, pp 453-477 (together with Ines Verena Arnolds).

Furthermore, in the service industry, customer classes sometimes have to be defined first by being as  homogeneous as possible within a group of customers (e.g., age group, buying behavior or product preferences). However, customers of different groups should be as different as possible. In various projects, I have developed clustering models that define customer groups and pursue different goals.

  • „Length of stay outlier detection in pediatrics.” International Conference on Information Systems (ICIS), Dublin, 2016. (together with Rema Padman, PhD)
  • “Workload Reduction Through Usability Improvement of Hospital Information Systems – The Case of Order Set Optimization. International Conference on Information Systems (ICIS)”, Fort Worth, TX, 2015. (together with Rema Padman, PhD and Yiye Zhang, PhD)
  • “Mathematical programming for cognitive workload reduction in order set optimization.” Health Care Management Science, 2018. (together with Rema Padman, PhD and Yiye Zhang, PhD)

Customer demands, which at the operational level, for example, yield an uncertain profit for a company, can be predicted with classification methods. A paper published in the INFORMS Journal on Computing addresses a problem in which the profit function of customers is classified by a machine learning method. Then, based on the customer's assignment to one of more than 500 different classes, a resource allocation decision is made. In a further work it could be shown that methods of machine learning to predict no-shows can give very good results. This is especially important in the provision of services where the resource is very expensive and one wants to minimize opportunity costs.

  • “Machine Learning Approaches for Early DRG Classification and Resource Allocation”. INFORMS Journal on Computing, 27(4), 2015. (together with Prof. Dr. Rainer Kolisch, Daniel B. Neill, PhD and Rema Padman, PhD)

Research Group

Current supervision

Emma Aspland

Emma Aspland

Research student

Sarie Brice

Research student

Mark Tuson photograph

Mark Tuson

Research student

John Threllfall

John Threlfall

Research student

Emily Williams

Emily Williams

Research student