Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Forecasting Ferry Passenger Flow Using Long-Short Term Memory Neural Networks

Daniel Fesalbon

TL;DR

The paper addresses forecasting ferry passenger flow for two Philippine ports using a two-layer LSTM-based time-series model. It utilizes monthly passenger data from 2016–2022 provided by the Philippine Ports Authority, trained with Adam optimization and evaluated primarily by $MAPE$, achieving averages of $0.28$ for Batangas and $0.26$ for Mindoro. The study demonstrates that the LSTM approach can produce reasonable forecasts and offers a basis for planning maritime operations, while suggesting comparisons with alternative methods and extensions to additional ports. Overall, the work highlights the practicality of deep learning for port-level passenger forecasting and sets the stage for further methodological refinements.

Abstract

With recent studies related to Neural Networks being used on different forecasting and time series investigations, this study aims to expand these contexts to ferry passenger traffic. The primary objective of the study is to investigate and evaluate an LSTM-based Neural Networks' capability to forecast ferry passengers of two ports in the Philippines. The proposed model's fitting and evaluation of the passenger flow forecasting of the two ports is based on monthly passenger traffic from 2016 to 2022 data that was acquired from the Philippine Ports Authority (PPA). This work uses Mean Absolute Percentage Error (MAPE) as its primary metric to evaluate the model's forecasting capability. The proposed LSTM-based Neural Networks model achieved 72% forecasting accuracy to the Batangas port ferry passenger data and 74% forecasting accuracy to the Mindoro port ferry passenger data. Using Keras and Scikit-learn Python libraries, this work concludes a reasonable forecasting performance of the presented LSTM model. Aside from these notable findings, this study also recommends further investigation and studies on employing other statistical, machine learning, and deep learning methods on forecasting ferry passenger flows.

Forecasting Ferry Passenger Flow Using Long-Short Term Memory Neural Networks

TL;DR

The paper addresses forecasting ferry passenger flow for two Philippine ports using a two-layer LSTM-based time-series model. It utilizes monthly passenger data from 2016–2022 provided by the Philippine Ports Authority, trained with Adam optimization and evaluated primarily by , achieving averages of for Batangas and for Mindoro. The study demonstrates that the LSTM approach can produce reasonable forecasts and offers a basis for planning maritime operations, while suggesting comparisons with alternative methods and extensions to additional ports. Overall, the work highlights the practicality of deep learning for port-level passenger forecasting and sets the stage for further methodological refinements.

Abstract

With recent studies related to Neural Networks being used on different forecasting and time series investigations, this study aims to expand these contexts to ferry passenger traffic. The primary objective of the study is to investigate and evaluate an LSTM-based Neural Networks' capability to forecast ferry passengers of two ports in the Philippines. The proposed model's fitting and evaluation of the passenger flow forecasting of the two ports is based on monthly passenger traffic from 2016 to 2022 data that was acquired from the Philippine Ports Authority (PPA). This work uses Mean Absolute Percentage Error (MAPE) as its primary metric to evaluate the model's forecasting capability. The proposed LSTM-based Neural Networks model achieved 72% forecasting accuracy to the Batangas port ferry passenger data and 74% forecasting accuracy to the Mindoro port ferry passenger data. Using Keras and Scikit-learn Python libraries, this work concludes a reasonable forecasting performance of the presented LSTM model. Aside from these notable findings, this study also recommends further investigation and studies on employing other statistical, machine learning, and deep learning methods on forecasting ferry passenger flows.
Paper Structure (11 sections, 7 equations, 4 figures, 4 tables)

This paper contains 11 sections, 7 equations, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Studies
  3. Methods
  4. Long-Short Term Memory Recurrent Neural Networks
  5. Data Acquisition and Processing
  6. Evaluation Metrics
  7. Results
  8. LSTM Model Specification
  9. Ferry Passenger Data
  10. Model Training and Forecasting Performance
  11. Conclusions

Figures (4)

  • Figure 1: Illustrates the structure of the LSTM memory cell andréassond2020 consisting of four elements such as forget gate (ft), input gate (it), reconnection neuron ($\tilde{C\textsubscript{t}}$), and the output gate (ot).
  • Figure 2: LSTM Model's Forecast/Prediction on Port Mindoro Monthly Passenger Test Sets
  • Figure 3: Batangas Port Monthly Passenger Traffic from 2016 up to 2022
  • Figure 5: LSTM Model's Forecast/Prediction on Port Batangas Monthly Passenger Test Sets