Table of Contents

1. Project Overview
   • Brief
   • Aim
   • Scope
2. Features
   • Crane Optimization
   • Machine Learning
3. Problem Design
   • Scenario
4. Logical Representation
5. Transitional Model
6. Search Algorithm
   • Algorithm
7. Machine Learning Model
   • Features

Project Overview

Brief

This project model solves a complex crane optimization problem and trains a machine learning model to predict container priority when placing.

Aim

To explore AI solutions for automation combining optimized algorithms and supervised learning

Scope

Includes problem design, logical modeeling, search algorithms, machine learning and ethical reflections which is the priority

Features

Crane Optimization

• Logical relations to represent crane environments
• Transition models for crane actions
• Search algorithm implementation for finding an optimal crane plan

Machine Learning

• Predicting container priorities using supervised learning
• Training and evaluation with real-world data

Problem Design

Scenario

• 6 Containers, 4-8 loading bays
• At least 10 crane actions to solve the problem Goal: Define and Achieve an optimal crane configuration through logical planning

Logical Representation

Relationed Defined: SOON

Goals: Ensure coverage of the environment and actions

Transitional Model

• Preconditions and effects for each action
• Documented in a transition table

Search Algorithm

Algorithm

Breadth-first search (BFS) A*

• Does it find a solution
• Does it guarantee the best solution?

Machine Learning Model

• Dataset: ContainerData.csv
• Features: Height, Width, Movement Frequency, Dock time
• Goal: Train a model to predict container priority (high/low).

Appendix

State 1

State 2

State 3

State 4

State 5