Predicting Commercial Vehicle Parking Duration using Generative Adversarial Multiple Imputation Networks

Abstract

<jats:p> As the world rapidly urbanizes in pace with economic growth, the rising demand for products and services in cities is putting a strain on the existing road infrastructure, leading to traffic congestion and other negative externalities. To mitigate the impacts of freight movement within commercial areas, city planners have begun focusing their attention on the parking behaviors of commercial vehicles. Unfortunately, there is a general lack of information on such activities because of the heterogeneity of practices and the complex nature of urban goods movement. Furthermore, field surveys and observations of truck parking behavior are often faced with significant challenges, resulting in the collection of sparse and incomplete data. The objective of this study is to develop a regression model to predict the parking duration of commercial vehicles at the loading bays of retail malls and identify significant factors that contribute to this dwell time. The dataset used in this study originates from a truck parking and observation survey conducted at the loading bays of nine retail malls in Singapore, containing information about the trucks’ and drivers’ activities. However, because of the presence of incomplete fields found in the dataset, the authors propose the use of a generative adversarial multiple imputation networks algorithm to impute the incomplete fields before developing the regression model using the imputed dataset. Through the parking duration model, the activity type, parking location, and volume of goods delivered (or picked up) were identified as significant features influencing vehicle dwell time, corroborating with findings in the literature. </jats:p>

Estimation of Occupancy Using IoT Sensors and a Carbon Dioxide-Based Machine Learning Model with Ventilation System and Differential Pressure Data

“…The data used for learning were collected for 55 days—13 September to 6 November—and data that was not collected due to Internet errors was excluded. Low, R et al [ 33 ] and Stekhoven et al [ 34 ] were used various methods of interpolating missing data. However, in this paper, among the data from 13 September to 6 November used for learning, from 11:52 on 16 September to 13:57 on 21 September, the input values of differential pressure and CO2 concentration data were not measured due to an internet connection error.…”

Two-Step Approach for Occupancy Estimation in Intensive Care Units Based on Bayesian Optimization Techniques

“…In order to deal with missing data, a schema for input data was required to fill values on such empty variables. Although there are other imputation methods, such as MissForest [ 27 ] or generative adversial networks [ 28 ], in this work we have proposed using K–Nearest Neighbors, as it is one of the most widely used in the current state of the art, in addition to its simplicity in implementation and usage.…”

A Semi-Supervised Approach for Improving Generalization in Non-Intrusive Load Monitoring

“…However, due to their high performance in image processing, they were also exploited in other domains. For example, [ 51 ] utilized the GAN approach for data imputation in the transportation domain, whilst [ 52 ] reviewed generative models for the graph generation. This wide acceptance of GAN influenced further improvement in the field, such as with the DANN proposition.…”