Publicação
Statistical analysis of variables of tires' specifications and respective tests
| Resumo: | The objective of this study was to identify the most important predictor variables regarding the dimensions of an agricultural tire, as well as the development of a tool, using various methods, that would allow the user to predict the nal dimensional results, as well as understand the impact that each variable’s value change would have on that same result. In order to do that linear regression and logistic regression were used. To make the predictions about the nal dimensional results, two different models based on linear regression were developed; one for each of the response variables (X36 and X37) regarding the tire’s dimensions. This also allowed the understanding of which variables were the most impactful ones on the nal result for each of the dimensions. The results were then validated using values that weren’t present in the initial data, with the objective of understanding if the obtained models were correctly adjusted, and if the they were providing good simulations. After analyzing the results it was safe to conclude that the error of the simulations was within an acceptable range. On the other hand, the logistic regression was applied to understand, given the values of the variables, what was the chance of a tire passing or failing the dimensional test. Finally, in order to facilitate the calculation of the previously mentioned simulations for new tires, there was a need to develop a tool that would allow the user to obtain results in a friendly environment. This tool was developed in the R Shiny environment. All the modeling was made using the software R, version 3.5.2. (2018-12-20). |
|---|---|
| Autores principais: | Santos, Renato Fernandes dos |
| Assunto: | Agricultural tire Dimensional results Linear regression Logistic regression Predictions R Shiny Variable impact Impacto das variáveis Pneu agrícola Previsões Regressão linear Regressão logística Resultados dimensionais |
| Ano: | 2019 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade do Minho |
| Idioma: | inglês |
| Origem: | RepositóriUM - Universidade do Minho |
| Resumo: | The objective of this study was to identify the most important predictor variables regarding the dimensions of an agricultural tire, as well as the development of a tool, using various methods, that would allow the user to predict the nal dimensional results, as well as understand the impact that each variable’s value change would have on that same result. In order to do that linear regression and logistic regression were used. To make the predictions about the nal dimensional results, two different models based on linear regression were developed; one for each of the response variables (X36 and X37) regarding the tire’s dimensions. This also allowed the understanding of which variables were the most impactful ones on the nal result for each of the dimensions. The results were then validated using values that weren’t present in the initial data, with the objective of understanding if the obtained models were correctly adjusted, and if the they were providing good simulations. After analyzing the results it was safe to conclude that the error of the simulations was within an acceptable range. On the other hand, the logistic regression was applied to understand, given the values of the variables, what was the chance of a tire passing or failing the dimensional test. Finally, in order to facilitate the calculation of the previously mentioned simulations for new tires, there was a need to develop a tool that would allow the user to obtain results in a friendly environment. This tool was developed in the R Shiny environment. All the modeling was made using the software R, version 3.5.2. (2018-12-20). |
|---|