Georg von GraevenitzABDULRAHMAN FAHAD ABDULRAHMAN ALLOAIHAN2022-05-292022-05-29https://drepo.sdl.edu.sa/handle/20.500.14154/46826This paper aim to reveal the results on a research study done to elaborate on how the use of artificial intelligence can improve telecommunication industries by reducing the number of customers terminating their contracts. The dataset being used for this paper has been collected from Kaggle and it belongs to a company located in the United States of America where the name of the company has not been released. As this paper is part of an academic nature, multiple machine learning models has been implemented to distinguish the differences between each model; such as, logistic regression, penalized logistic regression, random forest, decision tree, and multiple methods in survival analysis/regression. The aim of this project was to minimize the cost of transporting goods from the oil production factory to its final destination factoring in all possible routes that can be utilized from export nodes (factory) to import nodes (final destination) available. Moreover, each route has a per unit cost and has constraints; which are limitations on the amount of goods that can be exported from an export node to an import node with restrictions on the flow of minimum and maximum capacity allowed to go through each route. Optimizing the whole process is the ideal solution to deal with the increased demand because it will allow the transportation process to go smoothly and efficiently and will also allow the company to add various datasets that defer in constraints; such as, modify the number of routes available for each export node, modify the demand and supply of export and import nodes and finally the cost of transportation of goods. The objective of this project was to minimize the cost of the transportation process while taking into account the limitation of constraints given in each of the three different phases of the project. The phases were basic, intermediate and advanced and they were all solved and achieved using PULP library in python with an efficiency of 14 seconds, 18 seconds, and 19 seconds respectively for each phase.en