Blog 5 Data 621

Blog Entry 5: Predictive Analytics in Travel Planning

Title: Predictive Analytics in Travel Planning

Overview

In today’s fast-paced world, efficient travel planning is essential for individuals, businesses, and transportation authorities alike. Predictive analytics, particularly using machine learning models like Random Forest, can revolutionize travel planning by forecasting travel demand accurately. In this blog post, we explore how predictive analytics can enhance travel planning, using a Random Forest model as an example.

#Loading required libraries
#install.packages("randomForest")
library(randomForest)

## randomForest 4.7-1.1

## Type rfNews() to see new features/changes/bug fixes.

library(ggplot2)

## 
## Attaching package: 'ggplot2'

## The following object is masked from 'package:randomForest':
## 
##     margin

#Generating sample travel data
set.seed(123)
n <- 1000
travel_data <- data.frame(
  travel_date = seq(as.Date("2022-01-01"), by = "day", length.out = n),
  travel_demand = rpois(n, lambda = 50)
)

#Preprocess data
travel_data$month <- as.numeric(format(travel_data$travel_date, "%m"))
travel_data$day <- as.numeric(format(travel_data$travel_date, "%d"))
travel_data$weekday <- as.numeric(format(travel_data$travel_date, "%u"))
travel_data$is_weekend <- ifelse(travel_data$weekday %in% c(6, 7), 1, 0)

#Splitting data into training and testing sets
set.seed(456)
train_indices <- sample(1:nrow(travel_data), 0.8 * n)
train_data <- travel_data[train_indices, ]
test_data <- travel_data[-train_indices, ]

#Training Random Forest model
rf_model <- randomForest(travel_demand ~ month + day + weekday + is_weekend, data = train_data)

#Predicting travel demand for testing data
predictions <- predict(rf_model, newdata = test_data)

#Visualize predicted vs. actual travel demand
plot_data <- data.frame(
  Actual = test_data$travel_demand,
  Predicted = predictions
)

ggplot(plot_data, aes(x = Actual, y = Predicted)) +
  geom_point() +
  geom_abline(slope = 1, intercept = 0, color = "red", linetype = "dashed") +
  labs(x = "Actual Travel Demand", y = "Predicted Travel Demand", title = "Actual vs. Predicted Travel Demand")

Conclusion:

In conclusion, predictive analytics, powered by machine learning models like Random Forest, offers a powerful tool for enhancing travel planning. By leveraging historical data and relevant features, these models can provide valuable insights into future travel demand, enabling transportation agencies, businesses, and individuals to optimize resource allocation, improve operational efficiency, and deliver better customer experiences. As technology continues to advance, the integration of predictive analytics into travel planning processes will become increasingly essential for navigating the complexities of modern transportation systems.