Final Report Summary - MUSE-TECH (MUlti SEnsor Technology for management of food processes)
Executive Summary:
The basic concept of MUSE-Tech project is the integration of three High-End sensing technologies in a versatile Multi Sensor Device (MSD) for real-time monitoring of multiple parameters associated with the quality and the chemical safety of raw and in-process materials, to support Process Analytical Technology (PAT) implementation in the food industry, and to drive the food processes according to consistent levels of quality and chemical safety in the final products selected as case study (bread, chips and beer).
The main objectives of the project were:
- the development of three novel sensors based on state-of-the-art sensor technologies, i.e. Photoacoustic Spectroscopy (PAS), Quasi Imaging Vis-NIR (QIVN) and multipoint Diffuse Temperature Sensing (DTS), which will be tailored to on-line/in-line & real-time applications in the food industry in the context of PAT.
- the integration of the three novel sensors in a Multi Sensor Device (MSD), based on a versatile plug-in architecture, to monitor on-line/in-line & real-time several parameters associated with the quality and chemical safety of processed foods.
- to implement auto-adaptive software tools to automatically manage sensor data fusion, multivariate data analysis, predictive models and active adjustment of Critical Process Parameters.
Three MSDs have been assembled, one for each case study (bread, potato chips and beer). MSDs works by gathering information from the sensors about the characteristics of the raw and in-process materials, monitoring their key parameters and supporting, with the empirical predictive models, the adjustment of the process parameters.
The MUSE-Tech project reached some important objectives, demonstrating that novel sensors work quite properly under industrial conditions, and providing important statistical tools to model and adjust different food processes. In particular the DTS multipoint temperature sensor worked properly for the three case studies at both pilot plant and industrial level, and it has been proved to be enough reliable and effective to start the initial steps for its future exploitation and commercialization.
In bread production the MSD monitored flour composition and demonstrated to monitor simultaneously with great accuracy the temperature distribution as well as the humidity, CO2 and ethanol content in the atmosphere of the industrial dough prover.
In potato chips frying the MSD provided information about the temperature profiles in the frying oil tank, and monitored reducing sugar content in raw potatoes as well several parameters related with oil quality during frying (contamination with mineral oil, total polar matter, p-anisidine index, 2-4 decadienal).
The MSD could gather complex temperature profiles in the lauter tun and in the copper/brew kettle during beer production, as well as information about wort characteristics (extract, iodine number, alfa-aminonitrogen, CBB, glucose, fructose, maltose, maltoriose and dextrins).
Specific statistical treatments of the raw data, carried out by applying with the most recent statistical tools, and targeted software and user interfaces are other interesting findings of the project.
Finally, the mathematical models, developed to predict the quality of the final product based on data gathered on-line by the MSD during the process, were other interesting result for the potato chips and beer case studies.
By integrating three sensors, the MSDs were designed to monitor in real-time parameters associated with the quality and chemical safety of processed foods, so the main impact of the results could be on food equipment SME providers, which could offer innovative systems, and food producers of any size, but the concept of a multi sensor device is also eminently transferable to other production process such as the pharmaceutical and biotechnology industries.
Project Context and Objectives:
Variability introduced by the sequence of unit operations in food processing directly influences the compositional and sensorial properties as well as the safety of the final food products, hampering to meet consumer and retailer expectations. Conventional strategies of Quality Assurance can be effective, but are expensive and not flawless; batch failures and reworks are frequent. Unsuitable or noncompliant batches have to be discarded or reworked. For these reasons, food industry is trying to shift towards a novel holistic concept, namely “Quality by Design” (QbD) inspired by the PAT, which requires that quality of the food products should be incorporated into process development and design, not by post-production quality testing.
Successful implementation of the QbD and PAT approaches require tools and systems which enable continuous analysis and control of processes, including real-time (or nearly real-time) sensing and multivariate data analysis of both raw materials and in-process materials to assure achievement of end product quality specification at the completion of the process [Hussain 2002; FDA 2004; van den Berg et al., in press].
An ideal sensor for PAT applications in food operations should:
• be able to measure the parameters of interest with suitable reliability, accuracy and repeatability
• have the potential for a high measurement frequency, real-time measurements.
• be non-destructive, avoid sample handling and facilitate in-/on-line monitoring
• be easily interfaced with plant monitoring/control systems, easy-cleanable and stable
• be affordable and with low maintenance costs.
The main goals of MUSE-Tech are:
• to provide an affordable and versatile Multi Sensor Device, for on-line/in-line & real-time monitoring of multiple parameters associated with the quality and the chemical safety of the processed foods;
• to drive the food processes according to consistent levels of quality and chemical safety in the final products selected as case study (bread, chips and beer);
• to reduce manufacturing costs and reduce food waste through enabling flawlessly running of the processes.
• to disseminate the results to EU food chain stakeholders promoting market take-up of the technology.
The objectives of the project to reach these goals are:
• the development of three novel sensors based on state-of-the-art sensor technologies, i.e. Photoacoustic Spectroscopy (PAS), Quasi Imaging Vis-NIR (QIVN) and multipoint Diffuse Temperature Sensing (DTS), which will be tailored to on-line/in-line & real-time applications in the food industry in the context of PAT.
• the integration of the three novel sensors in a Multi Sensor Device (MSD), based on a versatile plug-in architecture, allowing additional inputs from sensors already installed in the process lines. The MSD will be able to monitor on-line/in-line & real-time a total of 28 parameters associated with the quality and chemical safety of processed foods (6 specific for bread production, 7 specific for chips frying, 11 specific for wort mashing and boiling and 4 general for the three case studies).
• to develop protocols and methodologies to pre-process the sensor signals and to improve the robustness and the performances of the real-time measurements in the three case studies;
• to build calibrations and empirical multivariate predictive models for each targeted quality and chemical safety attributes in the three case studies (bread production, chips and brewing);
• to implement auto-adaptive software tools to automatically manage sensor data fusion, multivariate data analysis, predictive models and active adjustment of CPPs for each case study;
• demonstration of the versatility and fitness for purpose of the Multi Sensor Device in three case studies (dough mixing and proving, potato chips frying and brewing) at industrial and pilot plant level.
• to establish a plan with specific measures (webpage, workshop organisation, newsletters, training activities, etc.) to disseminate the research outcomes to target end users, such as technological companies, food equipment producers, the European food industry, food research area and society at large.
Project Results:
See document attached: "S&T results/foreground"
Potential Impact:
See document attached: "Impact_Dissemination_Exploitation"
List of Websites:
Project website: www.musetech.eu
Coordinator: Dr. Massimo Castellari (massimo.castellari@irta.cat / +34-972630052)
Project Manager: Dr. Lluis Salvà (lluis.salva@irta.cat / +34-972630052)
The basic concept of MUSE-Tech project is the integration of three High-End sensing technologies in a versatile Multi Sensor Device (MSD) for real-time monitoring of multiple parameters associated with the quality and the chemical safety of raw and in-process materials, to support Process Analytical Technology (PAT) implementation in the food industry, and to drive the food processes according to consistent levels of quality and chemical safety in the final products selected as case study (bread, chips and beer).
The main objectives of the project were:
- the development of three novel sensors based on state-of-the-art sensor technologies, i.e. Photoacoustic Spectroscopy (PAS), Quasi Imaging Vis-NIR (QIVN) and multipoint Diffuse Temperature Sensing (DTS), which will be tailored to on-line/in-line & real-time applications in the food industry in the context of PAT.
- the integration of the three novel sensors in a Multi Sensor Device (MSD), based on a versatile plug-in architecture, to monitor on-line/in-line & real-time several parameters associated with the quality and chemical safety of processed foods.
- to implement auto-adaptive software tools to automatically manage sensor data fusion, multivariate data analysis, predictive models and active adjustment of Critical Process Parameters.
Three MSDs have been assembled, one for each case study (bread, potato chips and beer). MSDs works by gathering information from the sensors about the characteristics of the raw and in-process materials, monitoring their key parameters and supporting, with the empirical predictive models, the adjustment of the process parameters.
The MUSE-Tech project reached some important objectives, demonstrating that novel sensors work quite properly under industrial conditions, and providing important statistical tools to model and adjust different food processes. In particular the DTS multipoint temperature sensor worked properly for the three case studies at both pilot plant and industrial level, and it has been proved to be enough reliable and effective to start the initial steps for its future exploitation and commercialization.
In bread production the MSD monitored flour composition and demonstrated to monitor simultaneously with great accuracy the temperature distribution as well as the humidity, CO2 and ethanol content in the atmosphere of the industrial dough prover.
In potato chips frying the MSD provided information about the temperature profiles in the frying oil tank, and monitored reducing sugar content in raw potatoes as well several parameters related with oil quality during frying (contamination with mineral oil, total polar matter, p-anisidine index, 2-4 decadienal).
The MSD could gather complex temperature profiles in the lauter tun and in the copper/brew kettle during beer production, as well as information about wort characteristics (extract, iodine number, alfa-aminonitrogen, CBB, glucose, fructose, maltose, maltoriose and dextrins).
Specific statistical treatments of the raw data, carried out by applying with the most recent statistical tools, and targeted software and user interfaces are other interesting findings of the project.
Finally, the mathematical models, developed to predict the quality of the final product based on data gathered on-line by the MSD during the process, were other interesting result for the potato chips and beer case studies.
By integrating three sensors, the MSDs were designed to monitor in real-time parameters associated with the quality and chemical safety of processed foods, so the main impact of the results could be on food equipment SME providers, which could offer innovative systems, and food producers of any size, but the concept of a multi sensor device is also eminently transferable to other production process such as the pharmaceutical and biotechnology industries.
Project Context and Objectives:
Variability introduced by the sequence of unit operations in food processing directly influences the compositional and sensorial properties as well as the safety of the final food products, hampering to meet consumer and retailer expectations. Conventional strategies of Quality Assurance can be effective, but are expensive and not flawless; batch failures and reworks are frequent. Unsuitable or noncompliant batches have to be discarded or reworked. For these reasons, food industry is trying to shift towards a novel holistic concept, namely “Quality by Design” (QbD) inspired by the PAT, which requires that quality of the food products should be incorporated into process development and design, not by post-production quality testing.
Successful implementation of the QbD and PAT approaches require tools and systems which enable continuous analysis and control of processes, including real-time (or nearly real-time) sensing and multivariate data analysis of both raw materials and in-process materials to assure achievement of end product quality specification at the completion of the process [Hussain 2002; FDA 2004; van den Berg et al., in press].
An ideal sensor for PAT applications in food operations should:
• be able to measure the parameters of interest with suitable reliability, accuracy and repeatability
• have the potential for a high measurement frequency, real-time measurements.
• be non-destructive, avoid sample handling and facilitate in-/on-line monitoring
• be easily interfaced with plant monitoring/control systems, easy-cleanable and stable
• be affordable and with low maintenance costs.
The main goals of MUSE-Tech are:
• to provide an affordable and versatile Multi Sensor Device, for on-line/in-line & real-time monitoring of multiple parameters associated with the quality and the chemical safety of the processed foods;
• to drive the food processes according to consistent levels of quality and chemical safety in the final products selected as case study (bread, chips and beer);
• to reduce manufacturing costs and reduce food waste through enabling flawlessly running of the processes.
• to disseminate the results to EU food chain stakeholders promoting market take-up of the technology.
The objectives of the project to reach these goals are:
• the development of three novel sensors based on state-of-the-art sensor technologies, i.e. Photoacoustic Spectroscopy (PAS), Quasi Imaging Vis-NIR (QIVN) and multipoint Diffuse Temperature Sensing (DTS), which will be tailored to on-line/in-line & real-time applications in the food industry in the context of PAT.
• the integration of the three novel sensors in a Multi Sensor Device (MSD), based on a versatile plug-in architecture, allowing additional inputs from sensors already installed in the process lines. The MSD will be able to monitor on-line/in-line & real-time a total of 28 parameters associated with the quality and chemical safety of processed foods (6 specific for bread production, 7 specific for chips frying, 11 specific for wort mashing and boiling and 4 general for the three case studies).
• to develop protocols and methodologies to pre-process the sensor signals and to improve the robustness and the performances of the real-time measurements in the three case studies;
• to build calibrations and empirical multivariate predictive models for each targeted quality and chemical safety attributes in the three case studies (bread production, chips and brewing);
• to implement auto-adaptive software tools to automatically manage sensor data fusion, multivariate data analysis, predictive models and active adjustment of CPPs for each case study;
• demonstration of the versatility and fitness for purpose of the Multi Sensor Device in three case studies (dough mixing and proving, potato chips frying and brewing) at industrial and pilot plant level.
• to establish a plan with specific measures (webpage, workshop organisation, newsletters, training activities, etc.) to disseminate the research outcomes to target end users, such as technological companies, food equipment producers, the European food industry, food research area and society at large.
Project Results:
See document attached: "S&T results/foreground"
Potential Impact:
See document attached: "Impact_Dissemination_Exploitation"
List of Websites:
Project website: www.musetech.eu
Coordinator: Dr. Massimo Castellari (massimo.castellari@irta.cat / +34-972630052)
Project Manager: Dr. Lluis Salvà (lluis.salva@irta.cat / +34-972630052)
