Big Data in Transportation Engineering Nii Attoh-Okine Professor Department of Civil and Environmental Engineering University of Delaware, Newark, DE, USA Email: okine@udel.edu IEEE Workshop on Large Data Analytics in Transportation Engineering October 27, 2014 October 27, 2014 1 / 15
Big Data Big Data describes the collection of complex and large data sets such that it is difficult to capture, process, store, search and analyze using conventional data base systems. http://www.analytics-magazine.org/ October 27, 2014 2 / 15
Big Data October 27, 2014 3 / 15
5 V s of Big Data October 27, 2014 4 / 15
5 V s of Big Data Variety Structured and Semi-structured data. Sensor data Audio Video Combination of such data for efficient analysis. Velocity Increasing speed at which data is created, and the speed at which it can be processed, stored, and analyzed. E.g: Trade data collection, Pavement data collection, Traffic streams. October 27, 2014 5 / 15
5 V s of Big Data Volume It is defined as the amount of data generated in a period of time (e.g. Gigabytes, Terabytes) from different sources, locations to be stored and processed. Value Mainly refers to the worth of the information being managed. Veracity Refers to the trustworthiness of the data. It is an indication of the reliability and accuracy of information for making decisions. October 27, 2014 6 / 15
Economic Side Outline ticket purchasing Match market offers for passengers. Relevant offers for ticket agencies. Efficient feedback from customers and transportation agencies. Passing savings to customers. October 27, 2014 7 / 15
Travel Customers data to improve customer service - Airlines. Frequent flyers data may provide food choices, allergies, other preferences. October 27, 2014 8 / 15
Car, Truck, Train Monitor the condition of different parts (cars,trucks, trains). Train wheels. Travel patterns and routes. Delivery information. October 27, 2014 9 / 15
Predicting + Forecasting Transportation infrastructure performance and modeling. Modeling derailment of train based on rail defects data and geometry data. Travel forecasting. October 27, 2014 10 / 15
Example - Travel Data Analytics Applications I a) Anonymization b) Aggregation II III IV a) Interpretation b) Processing c) Modeling a) Time Patterns b) Spatial Signature c) Flows Patterns a) Operations b) Data Mining c) Strategies October 27, 2014 11 / 15
Issues Determining whether patterns from massive data represent noise or signal. Sampling errors are not consistent. More data - Error approach zero! October 27, 2014 12 / 15
Big Data Skills October 27, 2014 13 / 15
Research Improve individual tools Handle particular data types better. Make it easier to find entities in data. Make it easier to compose analyses from existing models. Improve the environment for exploring massive data? Pre-integrated data sets to provide context. Powerful infrastructure for data management and analytics. Rich collection of analytics and tools for analysis. Expertise in all aspects of the process. A great user experience through automation and intelligent guidance. Dr. Messatfa, IBM - Europe. October 27, 2014 14 / 15
Remarks October 27, 2014 15 / 15