• Outline ** Introduction - Problem To Be Addressed *** Purpose Or Objective Of Proposed Work
• The purpose is to introduce a simple method of determining a robust, optimal charging schedule for a fleet of Battery Electric Buses (BEBs) *** Literature Review, Including Theoretical Analysis (If Applicable)
• Pulling literature reviews/introductions from supporting documents

*** Significance And Scope Of Proposed Work

• Novel solution based off well studied problem
• Robust solution to allow for variation of real world events
• Accurate charge modeling due to nonlinear battery dynamics
• Quick results due to SA?
• Able to reduce peak hour usage
• Model includes other benefits "for free" due to the structure of the problem
  • Minimize chargers
  • Minimize consumption cost

** Background

• Berth Allocation Problem
• Position Allocation Problem
• Simulated Annealing
• Battery Dynamics (Linear/Nonlinear)

** Methods For Solving The Problem *** Plan For Accomplishing Objectives

• Modeling and simulation

*** Plan For Evaluating Results

• Compare results with existing MILP-PAP project

*** Timetable For Project Completion

• 1 month SA-PAP implementation completion

• 0.5-1 month battery dynamics

• Components Of Topics The following is a list of desired items to be included in the proposal.

• Position Allocation Problem (PAP) formulation

  • Built on from the Berth Allocation Problem (BAP)
  • Expand the formulation to not require information about the processing times

• Include Consumption/demand cost in the objective

  • Minimize the monetary cost of system by attempting to avoid peak hours
  • Minimize the monetary cost of system by reducing the peak15 of the system

• Nonlinear battery dynamics

  • More accurately predict the charging and discharging dynamics of LIBs

• Minimize chargers

  • Be able to predict what type of system setup is required for a set of routes

• Penalty method to allow for the bus to go below charge percentage

  • Having the charge be strictly above a certain percentage makes the solution space too small at times. This will allow the charges to vary more without killing the solver

• Simulated Annealing (SA) implementation

  • Pseudo code for the implementation
  • Written in rust (🦀)