Loading...
Publication
Aferição das diferenças de acuidade visual de rolhas de cortiça
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 91.32 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Portugal dis8ngue-se como o maior produtor de cor8ça, destacando-se na produção de
rolhas de cor8ça u8lizadas para vedação de vinho. A qualidade destas é avaliada por
classificação visual, uma técnica com elevada subje8vidade, exigindo assim a necessidade de
calibrar os analistas, de modo a garan8r uma maior uniformidade e fiabilidade dos resultados.
O estudo centrou-se na inves8gação das diferenças de acuidade visual e na capacidade
de deteção de defeitos em rolhas de cor8ça natural entre analistas da qualidade, avaliando a
repe8bilidade e reprodu8bilidade dos resultados, bem como a existência de diferenças
esta8s8camente significa8vas.
Foram analisadas 6 amostras de rolhas de diferentes classes e calibres por 15 analistas de
7 laboratórios dis8ntos. A análise dos dados ob8dos foi realizada através do método de Kapa de
Fleiss, do método Kapa de Cohen e do teste z.
Os resultados demonstraram uma boa repe8bilidade nas classificações realizadas pelos
mesmos analistas. No entanto, observaram-se variações entre as avaliações de analistas do
mesmo laboratório, em algumas amostras, sendo que essas diferenças foram esta8s8camente
significa8vas. Estas discrepâncias tornam-se ainda mais acentuadas na comparação entre
laboratórios, onde se verificou menor reprodu8bilidade e níveis de concordância reduzidos,
reforçando a necessidade de homogeneização dos critérios de classificação visual entre
empresas e de treino sistemá8co para reduzir as diferenças.
Portugal stands out as the largest cork producer, par8cularly in the manufacturing of cork stoppers used for wine sealing. The quality of these stoppers is assessed through visual inspec8on, a technique with high subjec8vity, thus requiring the calibra8on of analysts to ensure greater consistency and reliability of the results. The study focused on inves8ga8ng differences in visual acuity and the ability to detect defects in natural cork stoppers among quality control analysts, evalua8ng the repeatability and reproducibility of the results, as well as the existence of sta8s8cally significant differences. Six samples of cork stoppers of different grades and sizes were analyzed by 15 analysts from 7 different laboratories. The analysis of the obtained data was performed using Fleiss' Kapa method, Cohen's Kapa method, and the z-test. The results demonstrated good repeatability in the classifica8ons performed by the same analysts. However, varia8ons were observed among evalua8ons by analysts from the same laboratory, and in some samples, these differences were sta8s8cally significant. These discrepancies became even more pronounced when comparing results between laboratories, where lower reproducibility and reduced levels of agreement were observed, highligh8ng the need to harmonize visual classifica8on criteria across companies and systema8c training to reduce differences.
Portugal stands out as the largest cork producer, par8cularly in the manufacturing of cork stoppers used for wine sealing. The quality of these stoppers is assessed through visual inspec8on, a technique with high subjec8vity, thus requiring the calibra8on of analysts to ensure greater consistency and reliability of the results. The study focused on inves8ga8ng differences in visual acuity and the ability to detect defects in natural cork stoppers among quality control analysts, evalua8ng the repeatability and reproducibility of the results, as well as the existence of sta8s8cally significant differences. Six samples of cork stoppers of different grades and sizes were analyzed by 15 analysts from 7 different laboratories. The analysis of the obtained data was performed using Fleiss' Kapa method, Cohen's Kapa method, and the z-test. The results demonstrated good repeatability in the classifica8ons performed by the same analysts. However, varia8ons were observed among evalua8ons by analysts from the same laboratory, and in some samples, these differences were sta8s8cally significant. These discrepancies became even more pronounced when comparing results between laboratories, where lower reproducibility and reduced levels of agreement were observed, highligh8ng the need to harmonize visual classifica8on criteria across companies and systema8c training to reduce differences.
Description
Keywords
Cortiça Rolha de cortiça natural Classe visual Defeitos Cortiça Rolha de cortiça natural Classe visual Defeitos