Optimizing Utilization of Detectors

Abstract

This work seeks to increase the expected intelligence value collected by optimizing the time on multiple tasks. The purpose of this thesis is to provide a quantifiable process to determine how much time should be allocated to each task sharing the same asset. This optimized expected time allocation is calculated by numerical analysis and Monte Carlo simulation. Numerical analysis determines the expectation by involving an integral and a joint probability density function for a range of rates. In this case, rates are the historical hailing by taxi passengers. Monte Carlo simulation determines the optimum time allocation of the asset by repeatedly running experiments to approximate the expectation of the random variables. This was deemed necessary to account for real-world uncertainties as applied to a taxi scenario. The taxi variables consist of hail rates of the passengers, the fare amount for the task, and how much time to pursue said fare. Accounting for the uncertainty in the hail rates was exhibited by using ranges and not given values. The relationship the rates of hails for the taxi from two passengers and the fare values gathered is important to utilizing the taxi to maximize the total fare collected.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 2016
Accession Number
AD1027719

Entities

People

  • Crystal R. Warrene

Organizations

  • Naval Postgraduate School

Tags

Communities of Interest

  • Human Systems
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Algorithms
  • Data Science
  • Detectors
  • Distribution Functions
  • Information Science
  • Integrals
  • Monte Carlo Method
  • Numerical Analysis
  • Passengers
  • Probability
  • Probability Density Functions
  • Probability Distributions
  • Random Variables
  • Simulations
  • Statistical Analysis
  • Statistics
  • United States Naval Academy

Readers

  • Aerospace Engineering
  • Aerospace logistics and air mobility.
  • Mathematical Modeling and Probability Theory.