Can Neural Network Solve Everything? Case Study Of Contradiction In Logistic Processes With Neural Network Optimisation

Abstract

Neural networks have become the most popular family of algorithms of machine learning in today’s world. Neural networks are a computational model inspired by the behavior observed in the human brain. They consist of a set of units called neurons that are connected to transmit signals. This paper has gone through a basic overview of both neural networks and linear programming methods and compares them. Giving both examples of their applications in logistic problems as well as their advantages and disadvantages in their different aspects. It has been shown that when working with linear restrictions, transport, logistics, and optimization issues are better dealt with using linear programming methods. However, in the case of non-linear restrictions or objective functions, these methods will not be feasible. Therefore neural networks provide a valid and beneficial alternative to solve this type of problems.