Simulation-based decision support framework for dynamic ambulance redeployment in Singapore

OBJECTIVE Dynamic ambulance redeployment policies tend to introduce much more flexibilities in improving ambulance resource allocation by capitalizing on the definite geospatial-temporal variations in ambulance demand patterns over the time-of-the-day and day-of-the-week effects. A novel modelling framework based on the Approximate Dynamic Programming (ADP) approach leveraging on a Discrete Events Simulation (DES) model for dynamic ambulance redeployment in Singapore is proposed in this paper. METHODS The study was based on the Singapore's national Emergency Medical Services (EMS) system. Based on a dataset comprising 216,973 valid incidents over a continuous two-years study period from 1 January 2011-31 December 2012, a DES model for the EMS system was developed. An ADP model based on linear value function approximations was then evaluated using the DES model via the temporal difference (TD) learning family of algorithms. The objective of the ADP model is to derive approximate optimal dynamic redeployment policies based on the primary outcome of ambulance coverage. RESULTS Considering an 8min response time threshold, an estimated 5% reduction in the proportion of calls that cannot be reached within the threshold (equivalent to approximately 8000 dispatches) was observed from the computational experiments. The study also revealed that the redeployment policies which are restricted within the same operational division could potentially result in a more promising response time performance. Furthermore, the best policy involved the combination of redeploying ambulances whenever they are released from service and that of relocating ambulances that are idle in bases. CONCLUSION This study demonstrated the successful application of an approximate modelling framework based on ADP that leverages upon a detailed DES model of the Singapore's EMS system to generate approximate optimal dynamic redeployment plans. Various policies and scenarios relevant to the Singapore EMS system were evaluated.

[1]  Matthew S. Maxwell,et al.  Approximate Dynamic Programming for Ambulance Redeployment , 2010, INFORMS J. Comput..

[2]  Matthew S. Maxwell,et al.  Ambulance redeployment: An approximate dynamic programming approach , 2009, Proceedings of the 2009 Winter Simulation Conference (WSC).

[3]  Richard S. Sutton,et al.  Learning to predict by the methods of temporal differences , 1988, Machine Learning.

[4]  Marcus Eng Hock Ong,et al.  Dynamic ambulance reallocation for the reduction of ambulance response times using system status management. , 2015, The American journal of emergency medicine.

[5]  J. Harvey,et al.  The Emergency Medical Service Systems Act of 1973. , 1974, JAMA.

[6]  K. Song,et al.  Association between resuscitation time interval at the scene and neurological outcome after out-of-hospital cardiac arrest in two Asian cities. , 2014, Resuscitation.

[7]  Michel Gendreau,et al.  The maximal expected coverage relocation problem for emergency vehicles , 2006, J. Oper. Res. Soc..

[8]  P. Pons,et al.  Paramedic response time: does it affect patient survival? , 2005, Academic emergency medicine : official journal of the Society for Academic Emergency Medicine.

[9]  S. Weiss,et al.  Does Ambulance Response Time Influence Patient Condition among Patients with Specific Medical and Trauma Emergencies? , 2013, Southern medical journal.

[10]  Longxin Lin.  Self-Improving Reactive Agents Based on Reinforcement Learning, Planning and Teaching , 2004, Machine Learning.

[11]  Lark,et al.  OUTCOMES OF RAPID DEFIBRILLATION BY SECURITY OFFICERS AFTER CARDIAC ARREST IN CASINOS , 2000 .

[12]  W. David Kelton,et al.  Optimization model selection for simulation-based approximate dynamic programming approaches in semiconductor manufacturing operations , 2012, Proceedings Title: Proceedings of the 2012 Winter Simulation Conference (WSC).

[13]  Marcus Eng Hock Ong,et al.  Factors affecting the ambulance response times of trauma incidents in Singapore. , 2015, Accident; analysis and prevention.

[14]  Dean Uyeno,et al.  A practical methodology for ambulance location , 1984 .

[15]  George A. Wells,et al.  Advanced Cardiac Life Support in Out-of-Hospital Cardiac Arrest , 2004 .

[16]  Verena Schmid,et al.  Solving the dynamic ambulance relocation and dispatching problem using approximate dynamic programming , 2012, Eur. J. Oper. Res..

[17]  Sondes Hammami,et al.  A stochastic programming model for ambulance location allocation problem in the Tunisian context , 2013, Proceedings of 2013 International Conference on Industrial Engineering and Systems Management (IESM).

[18]  Chun-Ling Chuang,et al.  A MAXIMUM EXPECTED COVERING MODEL FOR AN AMBULANCE LOCATION PROBLEM , 2007 .

[19]  Gilbert Laporte,et al.  Application of the Double Standard Model for Ambulance Location , 2009 .

[20]  Peter J. Kolesar,et al.  ANNIVERSARY ARTICLE: Improving Emergency Responsiveness with Management Science , 2004, Manag. Sci..

[21]  Samir Elhedhli,et al.  A stochastic optimization model for real-time ambulance redeployment , 2013, Comput. Oper. Res..

[22]  Gilbert Laporte,et al.  Ambulance location and relocation models , 2000, Eur. J. Oper. Res..

[23]  Stephen Dean.  The origins of system status management. , 2004, Emergency medical services.

[24]  P. Pons,et al.  Selected Topics : Prehospital Care EIGHT MINUTES OR LESS : DOES THE AMBULANCE RESPONSE TIME GUIDELINE IMPACT TRAUMA PATIENT OUTCOME ? , 2002 .

[25]  Cem Saydam,et al.  A multiperiod set covering location model for dynamic redeployment of ambulances , 2008, Comput. Oper. Res..

[26]  Yun Yuan,et al.  A Chance Constrained Programming Model for Reliable Emergency Vehicles Relocation Problem , 2013 .

[27]  A. Shapiro.  Monte Carlo Sampling Methods , 2003 .

[28]  M. Ong,et al.  Geographic-time distribution of ambulance calls in Singapore: utility of geographic information system in ambulance deployment (CARE 3). , 2009, Annals of the Academy of Medicine, Singapore.

[29]  Marcus Eng Hock Ong,et al.  Geographical variation in ambulance calls is associated with socioeconomic status. , 2012, Academic emergency medicine : official journal of the Society for Academic Emergency Medicine.

[30]  Yee S. Ng,et al.  Ambulance Deployment under Demand Uncertainty , 2015 .

[31]  E. Lerner,et al.  The Golden Hour : Scientific Fact or Medical ‘ ‘ Urban Legend ’ ’ ? , 2001 .

[32]  Charles ReVelle,et al.  Siting Ambulances and Fire Companies: New Tools for Planners , 1991 .

[33]  M. Ong,et al.  A Quantile Regression Analysis of Ambulance Response Time , 2013, Prehospital emergency care : official journal of the National Association of EMS Physicians and the National Association of State EMS Directors.

[34]  J. Hedges,et al.  Does out-of-hospital EMS time affect trauma survival? , 1995, The American journal of emergency medicine.

[35]  Dah-Ming Shiah,et al.  Ambulance allocation capacity model , 2007, 2007 9th International Conference on e-Health Networking, Application and Services.

[36]  H. Akaike,et al.  Information Theory and an Extension of the Maximum Likelihood Principle , 1973 .

[37]  Dan Zhang,et al.  An Approximate Dynamic Programming Approach to Network Revenue Management with Customer Choice , 2009, Transp. Sci..

[38]  H. Oh,et al.  Reducing Ambulance Response Times Using Discrete Event Simulation , 2014, Prehospital emergency care : official journal of the National Association of EMS Physicians and the National Association of State EMS Directors.

[39]  John N. Tsitsiklis,et al.  On Average Versus Discounted Reward Temporal-Difference Learning , 2002, Machine Learning.

[40]  Mateo Restrepo,et al.  Erlang loss models for the static deployment of ambulances , 2009, Health care management science.

[41]  J. Pell,et al.  Effect of reducing ambulance response times on deaths from out of hospital cardiac arrest: cohort study , 2001, BMJ : British Medical Journal.

[42]  Marcus Eng Hock Ong,et al.  Reducing ambulance response times using geospatial-time analysis of ambulance deployment. , 2010, Academic emergency medicine : official journal of the Society for Academic Emergency Medicine.

[43]  Loo Hay Lee,et al.  An approximate dynamic programming approach for the empty container allocation problem , 2007 .

[44]  Joseph S. Pliskin,et al.  A geographic information system simulation model of EMS: reducing ambulance response time. , 2004, The American journal of emergency medicine.

[45]  Christopher M. Bishop,et al.  Pattern Recognition and Machine Learning (Information Science and Statistics) , 2006 .

[46]  Kevin P. Hwang,et al.  Using a discrete-event simulation to balance ambulance availability and demand in static deployment systems. , 2009, Academic emergency medicine : official journal of the Society for Academic Emergency Medicine.

[47]  Mark S. Daskin,et al.  A Maximum Expected Covering Location Model: Formulation, Properties and Heuristic Solution , 1983 .

[48]  E. T. Wilde,et al.  Do emergency medical system response times matter for health outcomes? , 2013, Health economics.

[49]  Richard L. Church,et al.  The maximal covering location problem , 1974 .

[50]  Andrew G. Barto,et al.  Linear Least-Squares Algorithms for Temporal Difference Learning , 2005, Machine Learning.