医疗服务系统到达过程的建模与预测
|
摘要
在服务运营管理研究领域,需求可变性的存在,对服务质量有着巨大的影响。医疗服务系统作为服务运营管理领域的重要代表,如何准确地拟合及预测病人的到达分布,对于医院合理安排资源配置起到了至关重要的作用。以以色列一家医院2007年1-10月份的急诊病人到达数据为基础,建立不同模型——非齐次泊松(NHPP)模型与幂律分布(Power-Law)模型——刻画急诊部门(ED)病人的到达规律,结果表明NHPP模型对数据的检验及拟合效果较好而幂律分布则仅能准确地刻画部分到达数据的特征。而为了进一步探究模型的适用性,对NHPP模型进行仿真,用幂律分布模型检验NHPP模型仿真得到的数据,结果显示幂律分布模型也能够较好地刻画该数据的分布特征,即印证了在一定条件下不同模型之间的可转化性,也为解决研究问题的多样化方法提供了有益思考。
In service operations management field,it is well known that the variability exists in the systems and has a huge impact on service quality.For healthcare system,which is a typical example of service systems,the ability of accurate fitting and predicting the distribution of its arriving data,that is,the variability of the data,is critical to effective management of resource planning.Therefore,based on Israel hospital data,two different models—NHPP model and Power-Law model – were built to study the arrival pattern of patients in emergency department.The results show that NHPP model behaves well while Power-Law model is just partially suitable when fitting the data.Also,in order to further explore the applicability of the models,Power-Law model was used to fit and test the simulated data from NHPP model,and the result is significant.This finding suggests that different models have some inner connections and inspires us to think of different methods and approaches to solve the same problems.
In service operations management field,it is well known that the variability exists in the systems and has a huge impact on service quality.For healthcare system,which is a typical example of service systems,the ability of accurate fitting and predicting the distribution of its arriving data,that is,the variability of the data,is critical to effective management of resource planning.Therefore,based on Israel hospital data,two different models—NHPP model and Power-Law model – were built to study the arrival pattern of patients in emergency department.The results show that NHPP model behaves well while Power-Law model is just partially suitable when fitting the data.Also,in order to further explore the applicability of the models,Power-Law model was used to fit and test the simulated data from NHPP model,and the result is significant.This finding suggests that different models have some inner connections and inspires us to think of different methods and approaches to solve the same problems.
引文
[1] Armony M,Israelit S,Mandelbaum A,et al.Patient flow in hospitals:A data-based queueing-science perspective[R].Working Paper,New York University,2011.
[2] Green V L.Capacity planning and management in hospitals[J].Operations Research and Health Care,2004,Springer,15-41.
[3] Green V L,Nguyen V.Strategies for cutting hospital beds:the impact on patient service[J].Health Services Research,2001,36(2):421-442.
[4] Kim S H,Whitt W.Statistical analysis with Little’s law[J].Operations Research,2013,61(4):1030-1045.
[5] Malmgren R D,Stouffer D B,Motter A E,et al.A Poissonian explanation for heavy tails in e-mail communication[J].PNAS,2008,105(47):18153-18158.
[6] Barabási A L.The origin of bursts and heavy tails in human dynamics[J].Nature,2005,435(5):207-211.
[7] Johansen A.Probing human response times[J].Physica A,2004,338(1):286-291.
[8] Bimpikis K,Markakis M G.Inventory pooling under heavy-tailed demand[J].Management Science,2015,62(6):1800-1813.
[9] Kim S H,Whitt W.Are Call Center and Hospital Arrivals Well Modeled by Nonhomogeneous Poisson Processes?[J].INFORMS,2014,16(3):464-480.
[10] Clauset A,Shalizi C R,Newman M E.J.Power-Law Distributions in Empirical Data[J].SIAM Review,2009,51(4):661-703.
[11] 宁如云.非齐次泊松过程的仿真方法[J].高等数学研究,2012,15(1):86-89.
[2] Green V L.Capacity planning and management in hospitals[J].Operations Research and Health Care,2004,Springer,15-41.
[3] Green V L,Nguyen V.Strategies for cutting hospital beds:the impact on patient service[J].Health Services Research,2001,36(2):421-442.
[4] Kim S H,Whitt W.Statistical analysis with Little’s law[J].Operations Research,2013,61(4):1030-1045.
[5] Malmgren R D,Stouffer D B,Motter A E,et al.A Poissonian explanation for heavy tails in e-mail communication[J].PNAS,2008,105(47):18153-18158.
[6] Barabási A L.The origin of bursts and heavy tails in human dynamics[J].Nature,2005,435(5):207-211.
[7] Johansen A.Probing human response times[J].Physica A,2004,338(1):286-291.
[8] Bimpikis K,Markakis M G.Inventory pooling under heavy-tailed demand[J].Management Science,2015,62(6):1800-1813.
[9] Kim S H,Whitt W.Are Call Center and Hospital Arrivals Well Modeled by Nonhomogeneous Poisson Processes?[J].INFORMS,2014,16(3):464-480.
[10] Clauset A,Shalizi C R,Newman M E.J.Power-Law Distributions in Empirical Data[J].SIAM Review,2009,51(4):661-703.
[11] 宁如云.非齐次泊松过程的仿真方法[J].高等数学研究,2012,15(1):86-89.