iHELP consists of the following modules that are programmed in Python:

1. Logistic Optimization and Simulation

2. User Interface for Data Input and Scenario Generation

3. Logistics Dashboard and Animation

4. Social Media Analytics

5. Weather Analytics

6. Personnel Management

1- Logistic Optimization and Simulation:

This is the core module that generates a logistical plan in a crisis situation. The python-based module, allocates the different types of items and resources to various available transporters of different modes and finds the best route to deliver them using Google API. The core also simulates the transporter movements using python’s DEVS for Discrete Event Simulation. The core module takes into account several constraints including transporter capacity, item transportability and item compatibility. The planning algorithm takes into account the urgency of the emergency needs as well as the efficiency of transporting items in the fastest possible way.

The route of the loaded transporter is refined by using a detailed navigation algorithm that calculates the travel time using Google API, while considering road closure and other disruptions including disruption due to weather conditions.

2- User Interface Module for Data Input and Scenario Generation

Logistics analysts from various organizations can input scenario data through a web interface. Data includes (i) disaster region that can be selected via a map interface (ii) needs requests in the affected areas including data extracted from social media platforms (iii) item types, reusable and consumable resources and their amounts to be transported to regions affected by a disaster.

3- Logistics Dashboard and Animation Module

Simulated logistical plans are visualized on a webpage. Road transporters and trains are projected on a google map that uses REACT technology. Air transporters and vessels are also overlayed on the same map to visualize a holistic plan.

4- Social Media Analytics Module

In many disasters, people resort to social media platforms to ask for help especially as Emergency Response Organizations get overwhelmed. In this module, text and image information is extracted from social media outlets such as Twitter and Facebook, evaluated for trustworthiness and relevance a disaster at a given location, and aggregated to be presented to decision makers.
Artificial Intelligence (AI) Natural Language Process (NLP) is used to extract disaster text information for different categories of incidents (need for food., collapsed building and roads, injuries, etc) and classify them as needs.
Image recognition and classification using neural network models are used to classify images that are (ir)relevant to a disaster and classfy them according to the type of incident. It also has the capability to detect whether an image was downloaded off the Internet or whether it is image taken by an eyewitness Both text and image are then combined and aggregated from many posts to a compute a trust score that reflects the type of incident in the disaster location and how trustworthy the information is. The posts as well as the aggregated information and computation are displayed on the frontend on Google map.

5- Weather Module

Weather API is used with iHELP to present the current and forecasted weather condition in order to reflect any extreme weather conditions on the logistical plan. The integrated weather module can be useful in determining if the logistical plan can be disrupted for iHELP to provide alternate plan and routes. The weather information is reflected on the frontend to display the current and forecasted condition around the disaster area

6- Personnel Management Module

This module includes staff and volunteers into the planning taking into account the skill type (doctors, nurses, firefighters, soldiers, etc). It takes into account schedule availability and other constraints. Unlike commercial workforce planning, in crisis management, personnel (staff and volunteers) have their own special constraints such as availability, transportability on various modes of transportation to move from one task to the next, skill level and other constraints.

iHELP – is a cloud-based simulator that is equipped with optimization algorithms for matching the needs of people at the various affected locations with the available resources and humanitarian relief. The combinations of commodities, people and other resources are then prioritized by algorithms that takes into account many factors including priorities, transportability, and compatibility of items to be transported as well as transporters utilization. iHELP then assigns transporters of various types and modes to different routes efficiently to reach their destinations in the shortest possible time (multi-commodity multi-mode transportation). This is a complex supply chain optimization problem and efficient algorithms are used to find good solutions even when the transportation network is disrupted.

Artificial Intelligence (AI) Natural Language Process (NLP) is used to extract disaster text information for different categories of incidents (flood, collapsed building and roads, injuries, etc) and classify them as needs.

Image recognition and classification using neural network models are used to classify images that are (ir)relevant to a disaster and classify them according to the type of incident. It also has the capability to detect whether an image was downloaded off the Internet or whether it is image taken by an eyewitness Both text and image are then combined and aggregated from many posts to a compute a trust score that reflects the type of incident in the disaster location and how trustworthy the information is. The posts as well as the aggregated information and computation are displayed on the frontend on Google map.