Wisit Kumphai, Ph.D., P.E., C.E.M., LEED AP

Algridas Bielskus

28-Sep-2017, 2:30-3:00PM

 
Session: The Art of Communicating Big Data to Decision Makers

Track M | Big Data and IoT


Smart City IoT Optimal Deployment Model through Needs-Based Geospatial and Financial Analyses

Wisit Kumphai, Ph.D., P.E., C.E.M., LEED AP
Sr. Energy & Sustainability Consultant
AECOM

Algridas Bielskus
Mechanical HVAC Engineer II
AECOM



The paper proposes and demonstrates a methodology to assess the needs and profitability of Smart City Internet of Things (IoT) devices using streetlight fixture platform, as well as determines the optimal deployment scenario to maximize overall city benefits from a streetlighting project. The optimal IoT feature selection is determined through qualitative and quantitative ranking analysis using Geospatial Information System (GIS). By analyzing the City of Chicago’s socioeconomic data, emergency evacuation routes, crime data, 311 call data, and other data sources, the modelling procedure determines each city neighborhood’s energy savings potential, social needs and revenue generation capabilities. The respective IoT features are then deployed in neighborhoods where a certain threshold of need or revenue generation capability is met. The GIS analysis allows custom weighting of social needs to allow the prioritization of IoT features that directly address key City goals and priorities. The paper describes streetlight deployment model using a mixed integer model linear optimization that balances energy savings, revenue generation, social need, and deployment duration. The backbone of the model is the quantitative social, revenue, and cost scores of each streetlight, derived from the GIS analysis. Each city neighborhood then has a certain deployment cost and social benefit when streetlights are deployed there. The optimization engine chooses which neighborhoods to deploy in every project year, subject to two constraints: Cashflow to the city shall remain positive at the end of each year and streetlights must be deployed in every neighborhood during the project. The Linear Simplex LP decision engine optimizes three separate city characteristics: Overall project cashflow, best social score to the city, and fastest fixture implementation time. The model then determines an overall best deployment scenario, with an equal weight for each of the city maximum characteristics.



 

Track: M Big Data and IoT | Session: The Art of Communicating Big Data to Decision Makers

 

Newsletter SignUp

 

No Spam - only latest news, program and activity updates!