IoT-Based quality monitoring system for banana beer

Abstract

The economic vision of Rwanda includes a “made in Rwanda” policy aimed at increasing economic competitiveness by enhancing Rwanda’s domestic market through value chain development. It is in line with this that Banana Beer Production (BBP) in Rwanda has been increasing day after day and spreading the drinking of locally made banana beer (Urwagwa) to the extent that demand for the beer has become high. However, there is very little monitoring of the quality of the beer produced. Implementation of methods to recognize the amount or level of parameters (like sugar, ethyl alcohol, methanol, acetic acid, and other volatile acid) that plays a major role in the beer quality is very scarce making the drinking of unsafe banana beer a real possibility. The intake of unsafe banana beer can cause harmful side effects such as burning of the mouth and throat, breathing difficulties, drooling, difficulty swallowing, stomach pain, vomiting and can also be a root cause of chronic diseases such as high blood pressure, blood sugar level, mental health problems and others. This tends to make BBP and banana beer consumption a public health concern in Rwanda. This master thesis research project focuses on prototyping the design and development of a Lowcost Banana Beer Alcohol Ingredient Detector (LBBAID) droppable into existing ready-made beer to assess the quality of the beer using the Internet of Things (IoT) and Machine Learning (ML) which is a branch of Artificial Intelligence (AI). The prediction of the quality of the beer can be done by observing patterns in the alcohol content and physicochemical properties of the drink. Our simulation results from Machine Learning show the best-fit ingredients and their amount in banana beer. AI, which is an effective non-linear multivariate tool in bioprocessing, with enormous generalization, prediction, and validation capabilities, is also compared with traditional optimization methods such as response surface methodology (RSM).