The selection of bed allocation methods is of practical significance for reducing the cost of bed mismatch in clinical units. By combining time series models for predicting patient admissions with the newsvendor model, we propose the core concept of “fitting error” anchored to the global optimal solution, and then derive the optimal number of bed configurations and the expected minimum bed mismatch cost. The newsvendor model based on time series forecasting not only breaks through the constraints of the classic newsvendor model, which adheres to specific distributions and single-period modeling, but also provides an “accurate portrait” of the “black-box nature” of regression residuals in time series. Numerical experiments of the model show that within a certain range of parameter values, decision-makers should choose the time series newsvendor model to configure beds; but outside this range of parameter values, decision-makers should choose the classic newsvendor model to configure beds. Empirical analysis also validates the results of the numerical experiments. Therefore, in the daily context of bed configuration, choosing the appropriate decision model is undoubtedly a more proactive and scientific application method.
为扩展研究结论在管理实践中的普适性,本文提出四类不同的时间序列模型:P阶自回归(Autoregressive of Order P,AR(P))模型,Q阶移动平均(Moving Average of Order Q,MA(Q))模型、P阶自回归Q阶移动平均(Autoregressive Moving Average of Order(P,Q),ARMA(P,Q))模型、指数平滑(Exponential Smoothing,ES)模型。后文将结合这四类时间序列模型对医院科室床位配置展开分析。
ZHAOL, WANGY X, ZHANGX Y, et al. Status and supply & demand analysis of hospital bed resources allocation in China[J]. Chinese Hospital Management, 2017, 37(8): 13-15 (Ch).
GUOS, JINGX L, LIQ, et al. Analysis on utilization efficiency of hospital beds in shandong province based on rank sum ratio[J]. Modern Hospital Management, 2020, 18(1): 41-44. DOI: 10.3969/j.issn.1672-4232. 2020.01.012(Ch ).
TANP F, CAIS H, FANZ Z, et al. Research on the allocation status and utilization efficiency of hospital bed resources in China[J]. Chinese Hospitals, 2023, 27(11): 21-24. DOI: 10.19660/j.issn.1671-0592.2023.11.05(Ch ).
[7]
HEB Y, DEXTERF, MACARIOA, et al. The timing of staffing decisions in hospital operating rooms: Incorporating workload heterogeneity into the newsvendor problem[J]. Manufacturing & Service Operations Management, 2012, 14(1): 99-114. DOI: 10.1287/msom.1110.0350 .
[8]
ALWANL C, XUM H, YAOD Q, et al. The dynamic newsvendor model with correlated demand[J]. Decision Sciences, 2016, 47(1): 11-30. DOI: 10.1111/deci.12171 .
[9]
SHENW W, LUOL, LUOL, et al. A data-driven newsvendor model for elective-emergency admission control under uncertain inpatient bed capacity[J]. Journal of Evidence‒Based Medicine, 2024, 17(1): 78-85. DOI: 10.1111/jebm.12599 .
XIONGH Q. An analysis of perishable product order timing in perspective of service level[J]. Industrial Engineering Journal, 2016, 19(3):24-29. DOI: 10.3969/j.issn.1007-7375.2016.03.005(Ch ).
SUZ H. Research on the influence of non-storability of service on service marketing and its counter-measures[J]. Inquiry into Economic Issues, 2012(02): 19-23. DOI: 10.3969/j.issn.1006-2912.2012.02.004(Ch ).
[14]
BAVAFAH, LEYSC M, ÖRMECIL, et al. Managing portfolio of elective surgical procedures: A multidimensional inverse newsvendor problem[J]. Operations Research, 2019, 67(6):1543-1563. DOI:10.1287/opre.2019.1848 .
CHENB, QIUM J, LINY B, et al. Feasibility of SARIMA model in predicting influenza-like illness cases in a tertiary hospital in Hainan [J]. Journal of Nanchang University(Medical Sciences), 2022, 62(2):75-78+99. DOI: 10.13764/j.cnki.ncdm.2022.02.015(Ch ).
[17]
褚宏睿.基于分布式鲁棒优化的医疗服务管理研究[D]. 北京: 北京理工大学, 2016.
[18]
CHUH R. Distributionally robust optimization with applications in healthcare service operations management[D]. Beijing: Beijing Institute of Technology, 2016(Ch).
[19]
OLIVARESM, TERWIESCHC, CASSORLAL. Structural estimation of the newsvendor model: An application to reserving operating room time[J]. Management Science, 2008, 54(1): 41-55. DOI:10.1287/mnsc.1070.0756 .
ZHOUH, HUS S, SHANJ N, et al. Temporal trends analysis of pediatric workload under the universal two-child policy background[J]. Chinese Hospital Management, 2021, 41(2): 47-51 (Ch).
[22]
ZHANGQ. Data science approaches to infectious disease surveillance[J]. Philosophical Transactions of the Royal Society A, 2022, 380(2214): 20210115. DOI: 10.1098/rsta.2021.0115 .
[23]
LUOL, XUX R, LIJ L, et al. Short-term forecasting of hospital discharge volume based on time series analysis[C]//2017 IEEE 19th International Conference on e-Health Networking, Applications and Services (Healthcom). New York: IEEE Press, 2017: 1-6. DOI: 10.1109/HealthCom.2017.8210801 .
[24]
HOSTEINSG, LARSENA, PACINOD, et al. A data-driven decision support tool to improve hospital bed cleaning logistics using discrete event simulation considering operators’ behaviour[J]. Operations Research for Health Care, 2023, 39: 100408. DOI: 10.1016/j.orhc.2023.100408 .
[25]
ZHUT, LIAOP, LUOL, et al. Data-driven models for capacity allocation of inpatient beds in a Chinese public hospital[J]. Computational and Mathematical Methods in Medicine, 2020, 2020: 8740457. DOI: 10.1155/2020/8740457 .
[26]
PIETERSA J, SCHLOBACHS. Combining process mining and time series forecasting to predict hospital bed occupancy[C]//International Conference on Health Information Science. Cham: Springer Nature, 2022: 76-87. DOI: 10.1007/978-3-031-20627-6_8 .
[27]
TANK W, NGQ Y, NGUYENF N H L, et al. Data-driven decision-support for process improvement through predictions of bed occupancy rates[C]//2019 IEEE 15th International Conference on Automation Science and Engineering (CASE). New York: IEEE Press, 2019: 133-139. DOI: 10.1109/COASE.2019.8843135 .
[28]
XUX R, LUOL, ZHONGX. Forecast-based newsvendor models for hospital bed capacity management[J]. IEEE Robotics and Automation Letters, 2021, 6(4): 6513-6520. DOI: 10.1109/LRA.2021.3093875 .
XUY J, XINL, LIUJ Q, et al. Research on the prediction of internet outpatient visits in public hospitals based on ARIMA and GM(1,1)model[J]. Modern Hospitals, 2024, 24(1): 14-19. DOI: 10.3969/j.issn.1671-332X.2024.01.005(Ch ).
[31]
KESHTKARL, SALIMIFARDK, FAGHIHN. A simulation optimization approach for resource allocation in an emergency department[J]. QScience Connect, 2015, 2015(1). DOI:10.5339/connect.2015.8 .
[32]
KAOE P C, TUNGG G. Bed allocation in a public health care delivery system[J]. Management Science, 1981, 27(5): 507-520. DOI: 10.1287/mnsc.27.5.507 .
[33]
KIMG, WUK, HUANGE. Optimal inventory control in a multi-period newsvendor problem with non-stationary demand[J]. Advanced Engineering Informatics, 2015, 29(1): 139-145. DOI: 10.1016/j.aei.2014.12.002 .
[34]
SHIY, ALWANL C, TANGC, et al. A newsvendor model with autocorrelated demand under a time-consistent dynamic CVaR measure[J]. IISE Transactions, 2019, 51(6): 653-671. DOI: 10.1080/24725854.2018.1539888 .
[35]
PUNIAS, SINGHS P, MADAANJ K. From predictive to prescriptive analytics: A data-driven multi-item newsvendor model[J]. Decision Support Systems, 2020, 136: 113340. DOI: 10.1016/j.dss.2020.113340 .
[36]
KAHNEMAND, TVERSKYA. Prospect theory: An analysis of decision under risk[J]. Econometrica, 1979, 47(2): 263. DOI: 10.2307/1914185 .
[37]
ANASTASIOUA, KARAGRIGORIOUA, KATSILEROSA. Comparative evaluation of goodness of fit tests for normal distribution using simulation and empirical data[J]. Biometrical Letters, 2020, 57(2): 237-251. DOI: 10.2478/bile-2020-0015 .