Article Details

Abstract

Crowd flow is an important operational indicator for public venues. The accurate prediction of crowd flow can improve operational efficiency, optimize resource allocation, and ensure public safety. In this study, SARIMA modeling is conducted on historical crowd flow time series data. The stationarity of the series is examined using the KPSS test to determine the differencing orders, STL decomposition is applied to the original series to analyze the seasonal characteristics of the data, and ACF and PACF analyses are combined to identify the seasonal and non-seasonal ARMA orders. Subsequently, a heat map is employed to visualize the AIC and BIC values, making the selection of the optimal model parameters more intuitive. Finally, the optimal model, SARIMA(2,1,5)(1,0,2)7, is obtained through training on 49 days of historical data. This model forecasts the daily crowd flow for the next 30 days, and the predictions show good agreement with the actual data. These results provide a useful reference for the management of public venues.

Keywords

crowd flow; SARIMA; KPSS; AIC; BIC

References

1. [1] Jiang H，Zhang X X L，Dong Y，et al. Data-Driven Predictive Modeling of Citywide Crowd Flow for Urban Safety Management: A Case Study of Beijing, China [J].Journal of Forecasting，2024，44（2）：730-752.
2. [2] Box G E P，Jenkins G M，Reinsel G C，et al. Time Series Analysis: Forecasting and Control [M].John Wiley & Sons，2015.
3. [3] Yule G U. VII. On a Method of Investigating Periodicities Disturbed Series, With Special Reference to Wolfer's Sunspot Numbers [J].Philosophical Transactions of the Royal Society，1927，226（636-646）：267-298.
4. [4] Wold H O A. On Prediction in Stationary Time Series [J].The Annals of Mathematical Statistics，1948，19（4）：558-567.
5. [5] Peychinov T，Karaivanova A，Mecheva T. Predicting Traffic Load Data: ARIMA and SARIMA Comparison [J/OL].Engineering Proceedings，2025，100（1）：29.https://doi.org/10.3390/engproc2025100029.
6. [6] Szostek K，Mazur D，Drałus G，et al. Analysis of the Effectiveness of ARIMA, SARIMA, and SVR Models in Time Series Forecasting: A Case Study of Wind Farm Energy Production [J/OL].Energies，2024，17（19）：4803.https://doi.org/10.3390/en17194803.
7. [7] Ma S Q，Liu Q Y，Zhang Y D. A prediction method of fire frequency: Based on the optimization of SARIMA model [J/OL].PLOS ONE，2021，16（8）：e0255857.https://doi.org/10.1371/journal.pone.0255857.
8. [8] HASSAN M M. Evaluating the Effectiveness of ARIMA and SARIMA Models for PM2.5 Forecasting in Bangladesh: A Time-Series Study (2000-2026) [PP/OL].https://www.researchsquare.com/article/rs-7348600/v1.
9. [9] Li Jie, Peng Qiyuan, Yang Yuxiang Passenger flow prediction of Guangzhou Zhuhai intercity railway based on SARIMA model [J]. Journal of Southwest Jiaotong University, 2020, 55 (1): 41-51
10. [10] Chen Zhilin and Hu Hongtao are welcoming each other Forecast of container throughput of Shanghai Port based on SARIMA model under the COVID-19 [J]. Industrial Engineering and Management, 2024, 29 (1): 32-40
11. [11] Deretić N，Stanimirović D，Awadh A M，et al. SARIMA Modelling Approach for Forecasting of Traffic Accidents [J/OL].Sustainability，2022，14（8）：4403.https://doi.org/10.3390/su14084403.
12. [12] Cheng Yanbiao, Zhao Qian, Tian Siyang, etc Application of SARIMA model in predicting the incidence of human brucellosis in Xinjiang Production and Construction Corps [J]. China Health Statistics, 2025, 42 (2): 282-285292
13. [13] Zhang Yueqin, Bai Lixia, Guangming, etc SARIMA model prediction of pertussis incidence in Shanxi Province [J]. China Health Statistics, 2024, 41 (4): 551-554
14. [14]Qiu H F，Zhao H，Xiang H Y，et al. Forecasting the Incidence of Mumps in Chongqing Based on a SARIMA Model [J/OL].BMC Public Health，2021，21（1）：373.https://doi.org/10.1186/s12889-021-10383-x.
15. [15] Guo Dan, Hu Bo, Liu Junde, etc Research on Monthly Power Generation Prediction of Wind Farm Based on SARIMA Model [J]. China Electric Power, 2016, 49 (2): 136-140
16. [16] Yin C L，Liu K H，Zhang Q J，et al. SARIMA-Based Medium- and Long-Term Load Forecasting [J].Strategic Planning for Energy and the Environment，2023，42（2）：283-306.
17. [17] Yang Zhenhao, Zhang Junbo, Yang Chenxing, etc Prediction of China's Consumer Price Index for Aquatic Products Based on SARIMA Model [J]. Marine Lake Bulletin, 2021, 43 (2): 131-138
18. [18] Ahadiyah K. Predicting Consumer Price Index Movements with SARIMA: A Case from Kediri City [J].Journal of International Business, Economics and Entrepreneurship，2025，10（2）：112-122.
19. [19] Chieu T Q，Hang V T K. Daily Stock Price Prediction: A Case Study in Vietnam [J].International Journal of Computer Science and Mobile Computing，2023，12（5）：17-31.
20. [20] Tamizharasi Dr，Purushottam Dr，Jahnavi Dr. Predicting Exchange Rate Between the us Dollar (USD) and Indian Rupee (INR): An Empirical Analysis Using Sarima Model [J].International Journal of Research in Finance and Management，2024，7（1）：56-63.
21. Zhao Qiang, Wang Qingyu, Shu Zhiguang Research on Short term Forecasting of Coastal Sea Surface Temperature Based on SARIMA Model [J]. Ocean Forecasting, 2024, 41 (1): 42-49
22. [22] Kabbilawsh P，Sathish D，R N. Forecasting Long-term Monthly Precipitation Using SARIMA Models [J/OL].Journal of Earth System Science，2022，131：174.https://www.ias.ac.in/article/fulltext/jess/131/0174.
23. [23] Ray S，Sankar S，Mishra P，et al. Time Series SARIMA Modelling and Forecasting of Monthly Rainfall and Temperature in the South Asian Countries [J].Earth Systems and Environment，2021，5（3）：531-546.

About the authors

Zhang Junjun (1991-), male, Han ethnicity, from Lianyuan, Hunan, senior engineer, master's student, research interests include time series analysis.