This study addresses the optimization of the bottling process in the wine industry, focusing particularly on the filtration stage, which accounts for approximately 30% of the total costs in the value chain. In the context of significant global production, but with challenges arising from declining demand and increasing costs, precise planning becomes necessary to ensure product quality and operational efficiency in order to control and minimize the costs associated with resource usage. The work proposes an operational approach based on a mixed-integer programming model that optimizes the sequencing of batches on the filtration and bottling lines, integrating setup times that depend on the processing order. An analysis of prevailing practices in the Chilean wine industry is conducted, complemented by a literature review that identifies knowledge gaps, particularly regarding the influence of the wine’s physicochemical properties and the potential for filter blockage. The results of the model allow for reduced processing times and improved batch allocation, highlighting the importance of considering hydraulic and operational variables for more precise planning. Furthermore, the study discusses the limitations of the current approach and suggests future directions to continue optimizing production processes in the wine industry.

The wine industry uses filtration systems to remove particles that could compromise the quality of wine before bottling. However, traditional systems present significant operational limitations, such as volume restrictions, intensive handling, and risks associated with high-pressure devices. This study addresses these challenges by proposing an innovative solution: a pneumatic pumping system that delivers a pulsed flow under constant pressure, replacing conventional pressurized tanks for clogging tests. This approach not only reduces analysis time by 66.2% but also improves operational safety and enables the development of new indicators that are not limited by tank volume. To validate this proposal, a robust methodological framework was developed, integrating techniques and methodologies typically addressed separately into a single comprehensive approach. This includes evaluating the reproducibility of measurements, performing a comparative analysis of hydraulic performance under standard conditions, quantifying the impact on processing times, and assessing user perceptions. This approach demonstrates that the proposed device contributes to faster operational decision-making without exposing workers to unnecessary risks while optimizing resource usage. It addresses a previously underexplored technical challenge and establishes a solid foundation for future research, highlighting its academic relevance and impact beyond a technical implementation.

The final filtration and bottling of wine represents the last and most important stage within the entire process of its production and commercialization. In this context, wine is filtered to retain particles and ensure its microbiological and organoleptic properties. The measurement of the Laurenty index during this process helps to prevent problems in flow and pressure, which can increase production costs and times. However, this index can only be measured when the wine is ready to be bottled, and in some cases, only a few hours before bottling, which limits its usefulness for proper production planning. To overcome these limitations, the use of Machine Learning is proposed to estimate the Laurenty index from 27 physicochemical parameters through an analysis of data over 30 weeks. For this purpose, missing data are imputed, dimensionality is reduced using principal component analysis (PCA) and factor analysis (FA), and models are trained and evaluated using the factors during the first 24 weeks. Finally, the model with the highest predictive power is selected and used to evaluate the subsequent 6 weeks. This is done by comparing the direct application of the model on a weekly basis to its retraining, considering the data generated in the immediately preceding week. Finally, it is concluded that retraining the model with the new data generated in the week prior to prediction ensures a high level of confidence, allowing its practical use at the industrial level to plan production.

The Chilean wine industry has evolved over decades, propelling Chile to become the leading exporter in Latin America and fourth globally, with revenues exceeding USD 2 billion annually. There are over 130,000 hectares dedicated to vine cultivation, yielding an annual wine production of 13 million hectoliters. Moreover, this industry contributes 0.5% to the GDP and generates over 100,000 direct jobs. However, the decrease in global wine consumption and the sustained increase in production costs in Chile have affected the profitability of this industry and have prioritized the control and reduction of operational costs. The final filtration of wine is one of the most critical stages within the production process, as it ensures the quality and microbiological stability of the wine during the bottling process. This stage is characterized by high uncertainty due to the nature, type, and concentration of particles that the wine may contain. To evaluate the particle content and its impact, comprehensive indicators are used based on physical analysis of samples through filtration tests on a laboratory scale, which requires trained technicians dedicated to conducting these analyses. Additionally, the high solids content in the wine during its elaboration only allows these methods to be used once the wine is clarified and ready for bottling. This does not allow for early production planning or may result in differences in productivity and costs compared to what was planned. This research aims to contribute to operational decision-making by using comprehensive indicators in the final wine filtration stage. To achieve this, the level of adoption of these methods in the Chilean industry is studied, the importance of their use for decision-makers is characterized, and a machine learning model is proposed to estimate their value throughout the production process.

The wine industry uses filtration systems to remove particles that could compromise the quality of wine before bottling. These systems, based on surface filtration media, trap particles that cause a clogging effect reducing the available area and decreasing the bottling flow rate. This clogging has a significant economic impact, as it deteriorates production times and increases operational costs, mainly due to higher labor demands. To obtain timely information on the performance of the filtering and bottling system, the industry uses indicators such as the maximum volume index, the filtration index, and the modified filtration index, which measure clogging potential. These are obtained through laboratory tests conducted at constant pressure under controlled conditions to replicate the industrial process. Traditional de-vices use pressurized tanks, where an inert gas at 30 psi pushes the fluid through a filter disc. However, these systems have disadvantages, such as requirement of intensive operational han-dling and the associated risks with high pressure devices. To address these challenges, a new constant-pressure system for wine filtration testing was de-veloped, utilizing a pneumatic pumping system without the need for a pressurized tank. This device, equipped with pressure regulators and hydraulic connections, reduces analysis time by 66.2% while improving operational safety. The validation of the device included statistical anal-yses showing that the samples processed by both devices were statistically similar. Additionally, a user survey revealed a high level of satisfaction with the new device compared to the traditional pressurized tank system. This approach has led to a tangible solution that is closely aligned with the specific needs of the industry, highlighting the practical application of scientific knowledge to address real industry challenges.