Open Access. Powered by Scholars. Published by Universities.®

Engineering Commons

Open Access. Powered by Scholars. Published by Universities.®

Civil and Environmental Engineering

LSU Master's Theses

Theses/Dissertations

Artificial neural networks

Publication Year

Articles 1 - 2 of 2

Full-Text Articles in Engineering

A Probabilistic Approach For Modeling And Real-Time Filtering Of Freeway Detector Data, Shourie Kondagari Jan 2006

A Probabilistic Approach For Modeling And Real-Time Filtering Of Freeway Detector Data, Shourie Kondagari

LSU Master's Theses

Traffic surveillance systems are a key component for providing information on traffic conditions and supporting traffic management functions. A large amount of data is currently collected from inductive loop detector systems in the form of three macroscopic traffic parameters (speed, volume and occupancy). Such information is vital to the successful implementation of transportation data warehouses and decision support systems. The quality of data is, however, affected by erroneous observations that result from malfunctioning or mis-calibration of detectors. The open literature shows that little effort has been made to establish procedures for screening traffic observations in real-time. This study presents a …


A Hybrid Model-Based And Memory-Based Short-Term Traffic Prediction System, Ciprian Danut Alecsandru Jan 2003

A Hybrid Model-Based And Memory-Based Short-Term Traffic Prediction System, Ciprian Danut Alecsandru

LSU Master's Theses

Short-term traffic forecasting capabilities on freeways and major arterials have received special attention in the past decade due primarily to their vital role in supporting various travelers' trip decisions and traffic management functions. This research presents a hybrid model-based and memory-based methodology to improve freeway traffic prediction performance. The proposed methodology integrates both approaches to strengthen predictions under both recurrent and non-recurrent conditions. The model-based approach relies on a combination of static and dynamic neural network architectures to achieve optimal prediction performance under various input and traffic condition settings. Concurrently, the memory-based component is derived from the data archival system …