title: "03-research-guidelines"

output: pdf_document

After surveying the related literature in the context of the MDRP, there are multiple solutions focused on solving the routing problems with varying degrees of complexity.

Understanding the problem is essential, and although all works have made a thorough effort in providing novel models and algorithms that solve the dynamic routing problem, these tend to focus on the model itself.

Businesses do not thrive as a consequence of applying the best or most complex algorithm.

On the contrary, companies succeed by understanding the operation and industry they belong to.

This leads to the opportunity and challenge of developing a holistic testing environment where the problem can be framed and analyzed, as well as providing a means to evaluate solution strategies.

Testing and measuring solutions is *as* important as their development, thus the following objective guides this research:

- Develop a computational framework where solutions for the Meal Delivery Routing Problem can be tested and gaps from the existing literature regarding operational dynamics are closed.

To achieve this objective, the following secondary objectives are stated:

1. Define operational dynamics missing from the existing MDRP literature.

1. Develop a discrete events simulator to represent the operation in which the MDRP functions.

1. Prepare real-life data instances to test solutions and feed the simulator.

1. Develop policies for matching orders to couriers.

1. Develop policies to capture the movement mechanics of couriers in the city.

1. Develop policies that model the acceptance or rejection of notifications by couriers.

1. Develop policies for modeling how orders are canceled by actors.

1. Design a computational framework where policies can be easily interchanged and operates with the simulator at its core.

1. Test different solution policies over a variety of scenarios using the proposed framework to obtain simulated performance metrics for policies comparison.