Publicação
Implementation of statistical process control in a bottling line in winery industry
| Resumo: | Introduction: The great demand of the markets has led to situations in which the production systems are characterized by the production of several batches but of small size. This new paradigm requires that adequate techniques be developed, both in terms of planning and in terms of Statistical Process Control (SPC), since there may be situations where it is not possible to collect enough data to properly estimate the parameters of the process (mean and variance). Objectives: Implementation of Statistical Process Control techniques in a wine industry in order to improve its final product. Methods: Whenever there is not enough data to properly estimate the parameters of the process, the suggested approach, is to adopt the developments proposed by Charles Quesenberry. In this case, the statistic of the sample at time i is transformed through the estimations of the process parameters using the information obtained (data) until the instant (i-1). The univariate study of process capability is performed through the capability indices QL and QU. Thus, in this paper, two situations of statistical control are addressed, one in which a univariate study is done, based on Q charts, and another in which the multivariate study is made, based on the MQ charts. Results: This study comprised the implementation of the SPC) of a wine brand that has low production volumes, in the process of bottling wine, which is considered a critical step since some care is needed, such as avoiding the occurrence of microbiological contaminations or oxidation of the wine. Conclusions: Whenever it is not possible to apply traditional control charts the use of the control Q charts (univariate analysis) and the control MQ charts (multivariate analysis) are the most appropriate choice not only for the control of one or more products but also for several sets of quality characteristics. |
|---|---|
| Autores principais: | Abreu, António |
| Outros Autores: | Requeijo, josé Gomes; Calado, João |
| Assunto: | Agriculture, Food and Veterinary Sciences |
| Ano: | 2018 |
| País: | Portugal |
| Tipo de documento: | artigo |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Instituto Politécnico de Viseu |
| Idioma: | inglês |
| Origem: | Millenium |
| Resumo: | Introduction: The great demand of the markets has led to situations in which the production systems are characterized by the production of several batches but of small size. This new paradigm requires that adequate techniques be developed, both in terms of planning and in terms of Statistical Process Control (SPC), since there may be situations where it is not possible to collect enough data to properly estimate the parameters of the process (mean and variance). Objectives: Implementation of Statistical Process Control techniques in a wine industry in order to improve its final product. Methods: Whenever there is not enough data to properly estimate the parameters of the process, the suggested approach, is to adopt the developments proposed by Charles Quesenberry. In this case, the statistic of the sample at time i is transformed through the estimations of the process parameters using the information obtained (data) until the instant (i-1). The univariate study of process capability is performed through the capability indices QL and QU. Thus, in this paper, two situations of statistical control are addressed, one in which a univariate study is done, based on Q charts, and another in which the multivariate study is made, based on the MQ charts. Results: This study comprised the implementation of the SPC) of a wine brand that has low production volumes, in the process of bottling wine, which is considered a critical step since some care is needed, such as avoiding the occurrence of microbiological contaminations or oxidation of the wine. Conclusions: Whenever it is not possible to apply traditional control charts the use of the control Q charts (univariate analysis) and the control MQ charts (multivariate analysis) are the most appropriate choice not only for the control of one or more products but also for several sets of quality characteristics. |
|---|