A Prophet-ARMA-based method for predicting outpatient visits in hospitals

Abstract

The invention discloses a Prophet-ARMA-based method for predicting outpatient visits in hospitals. The method comprises the following steps: 1, acquiring the outpatient data of the historical period adjacent to the period to be measured and generating the outpatient data set of each type of outpatient; 2, respectively inputting each outpatient data set into a Prophet model of each type of outpatient service to obtain first prediction data and fitting data; 3, calculating a residual sequence of each type of outpatient service; 4, respectively judging whether the residual sequence of each type of outpatient service is a pure random sequence, if not, inputting the residual sequence into the correspond ARMA model to obtain the second prediction data, and then executing the step 5; if yes, performing the step 6; 5, respectively adding the first prediction data and the second prediction data of the same type of outpatient clinics to obtain the prediction value of the amount of visit in the period to be tested; 6, taking the first prediction data as the prediction value of the attendance amount in the period to be measured. The invention combines Prophet and ARMA to improve the predictionaccuracy of the visiting quantity.

Description

technical field [0001] The invention belongs to the field of numerical modeling prediction, and in particular relates to a Prophet-ARMA prediction method for hospital outpatient visits. Background technique [0002] In recent years, it has been a common problem in large general hospitals for outpatients to seek medical treatment, which has exposed the contradiction between the unreasonable allocation of medical resources and the public's medical needs. There may be long queues in any service link of the outpatient clinic. How to improve the allocation efficiency of medical resources and effectively reduce the invalid waiting of outpatients has always been a concern of managers. Among the many hospital data indicators, outpatient visits have always been one of the important indicators to measure the quality of outpatient medical work. Scientifically analyze and predict the daily outpatient visits in hospitals, and analyze the changes and trend characteristics of the outpatien...

AI Technical Summary

Problems solved by technology

[0004] Among them, the number of hospital outpatient visits has long-term trend, seasonality, irregular periodicity such as holidays, and sequence characteristics such as some outliers. The existing Prophet method in the time series research and development method is only suitable for sequences with obvious periodicity or patterns. Time series characteristics, if the Prophet method is used to predict and fit the outpatient visits of hospitals, it is easy to miss the non-periodic stationary components of the target sequence, which reduces the accuracy of prediction

In addition, the ARMA method is used for forecasting, but the ARMA method is only suitable for short-term forecasting of stationary sequences, and several differential transformations are required to eliminate the sequence characteristics such as periodicity and seasonality of the sequence, resulting in large errors in the predicted sequence, while the stationary At the same time, for the period components with inconstant period length, such as the influence of holidays, the time intervals of holidays are not equal, and the ARMA model cannot strip the influence of holidays well. If it is applied to outpatient visits in hospitals quantity forecasting, the accuracy of the forecasting results is difficult to meet the requirements

Therefore, the existing time series research methods cannot meet the forecasting needs of hospital outpatient visits.

Images