TY - JOUR
T1 - Optimal staffing strategies for points of dispensing
AU - Hernandez, Ivan
AU - Ramirez-Marquez, Jose E.
AU - Starr, David
AU - McKay, Ryan
AU - Guthartz, Seth
AU - Motherwell, Matt
AU - Barcellona, Jessica
N1 - Publisher Copyright:
©2015 Elsevier Ltd. All rights reserved.
PY - 2015/5
Y1 - 2015/5
N2 - We present a heuristic-based multi-objective optimization approach for minimizing staff and maximizing throughput at Points-of-Dispensing (PODs). PODs are sites quickly set up by local health departments to rapidly dispense life-saving medical countermeasures during large-scale public health emergencies. Current modeling tools require decision makers to modify their models and re-run them for each "what if" scenario they are charged with preparing for, e.g. what happens if more/less staff are available. The exploration of these "what if" scenarios becomes tedious if there are many variables to change and the decision space quickly becomes too large to analyze effectively. Currently, to understand the trade-offs between throughput and staffing levels, public health emergency managers must maximize throughput subject to a specified staffing level. Then, they must repeatedly change the constraint (altering the maximum staff allowed) and re-run the model. In contrast, by approaching the problem from a multi-objective perspective and integrating discrete event and optimization tools, we automate of the exploration of the decision space. This approach allows public health emergency planners to examine far more potential solutions and to focus tangible planning resources on areas that show theoretical promise. Such an approach can also expose previously unidentified constraints in existing plans.
AB - We present a heuristic-based multi-objective optimization approach for minimizing staff and maximizing throughput at Points-of-Dispensing (PODs). PODs are sites quickly set up by local health departments to rapidly dispense life-saving medical countermeasures during large-scale public health emergencies. Current modeling tools require decision makers to modify their models and re-run them for each "what if" scenario they are charged with preparing for, e.g. what happens if more/less staff are available. The exploration of these "what if" scenarios becomes tedious if there are many variables to change and the decision space quickly becomes too large to analyze effectively. Currently, to understand the trade-offs between throughput and staffing levels, public health emergency managers must maximize throughput subject to a specified staffing level. Then, they must repeatedly change the constraint (altering the maximum staff allowed) and re-run the model. In contrast, by approaching the problem from a multi-objective perspective and integrating discrete event and optimization tools, we automate of the exploration of the decision space. This approach allows public health emergency planners to examine far more potential solutions and to focus tangible planning resources on areas that show theoretical promise. Such an approach can also expose previously unidentified constraints in existing plans.
KW - Multi-criteria analysis
KW - OR in health services
KW - Simulation
KW - Uncertainty modeling
UR - http://www.scopus.com/inward/record.url?scp=84924777806&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84924777806&partnerID=8YFLogxK
U2 - 10.1016/j.cie.2015.02.015
DO - 10.1016/j.cie.2015.02.015
M3 - Article
AN - SCOPUS:84924777806
SN - 0360-8352
VL - 83
SP - 172
EP - 183
JO - Computers and Industrial Engineering
JF - Computers and Industrial Engineering
ER -