Want to know more about this project?
Drop us a line!
Use Case:
FactorAI &
sustainability
The Digitalization of a Textile Company’s Production Process
in Support of Sustainability – by Andrea Spinelli, Industry BU Manager
Context
Digital transformation is revolutionizing the entire manufacturing sector, enabling improvements in efficiency, quality, and sustainability in line with the goals of Smart Manufacturing. The textile industry, in particular, is increasingly under scrutiny for its environmental impact, driven by water consumption, emissions, and waste production. However, investing in sustainable materials and the circular economy is not enough: for a true change, it is essential to optimize existing production processes and reduce waste.
FactorAI, Kode’s framework for Industry 5.0, enables real-time monitoring of the production cycle, ensuring both product quality and the reduction of costs, emissions, and waste.
In this case study, we analyze a project developed with FactorAI for a nonwoven fabric manufacturer, aimed at optimizing the production process in terms of cost and efficiency. Although the example pertains to the textile sector, the same challenges and opportunities apply to the entire manufacturing industry, demonstrating the value of Smart Manufacturing as a key to industrial sustainability.
The challenge
Like many of our clients, mid-sized manufacturing companies, this textile manufacturer already had machines equipped with sensors and connected to a centralized data collection system. However, the digitalization was only partial:
- some data were still stored in non-integrated formats (such as XLS, CSV, or non-integrated databases);
- sharing information between departments required time and resources;
- the system was limited to data storage without providing useful indicators for process analysis and optimization.
The system lacked in-depth key indicators that would allow for quick reading of progress data, nor did it have systems to identify corrective actions needed to make processes sustainable. It was necessary to identify the causes and remedies to reduce machine downtime, energy costs, and raw material consumption, while ensuring the quality of the final product.
Our Solution
Our Approach
To address these challenges, a structured approach is needed to transform simple data monitoring into an active tool for production process optimization. Thanks to FactorAI, the company can integrate, analyze, and leverage data to take real-time action and enhance operational sustainability.
Integration
The first step is to harmonize and centralize the company data in FactorAI. Since many connectors are typically already available, we can focus efforts on creating a specific acquisition system for data that are not yet integrated, such as Excel files and local databases.
Data Understanding
After integration, the next step is to normalize the information, eliminate redundancies, and structure the data for effective analysis. This enables the standardization of data from machines and different company systems, facilitating the generation of clear and automated reports.
The analysis focused on four key areas for sustainability:
- Energy consumption, to monitor the efficiency of each machine.
- Machine downtime and waste, identifying causes and reducing waste.
- Optimal machine usage, through the comparison of production parameters with ideal performance parameters.
- Product quality, correlating production data with quality standards.
These KPIs were essential for the continuous improvement of corporate sustainability.
Process control (MSPC)
Data collection, without effective analysis, has limited utility. For this reason, it is essential to develop an advanced dashboard that, through graphs, scatter plots, and control charts, enables:
- Real-time monitoring of key indicators;
- Correlating data from different sources to provide operational insights;
- Identifying relationships between production parameters and energy efficiency, optimizing machine settings.
The adoption of this technology makes it possible to:
- Provide a detailed comparison of energy consumption between machines and processes;
- Optimize production parameters to reduce waste and inefficiencies.
- Improve product quality through continuous monitoring of critical variables.
A central element of this phase is the Root-Cause Analysis, which helped identify the true origins of inefficiencies. This method enabled the company’s team to shift from a reactive to a proactive approach, implementing targeted solutions based on concrete data.
Predictive Analysis with AI
To ensure timely interventions and prevent inefficiencies, we developed Artificial Intelligence models focused on two key aspects:
- Predictive maintenance – Early detection of anomalies to prevent failures and reduce machine downtime.
- Production drift prevention– Models that flag real-time changes in production parameters that could lead to waste or quality defects.
These algorithms, based on machine learning, are trained on the company’s historical data to identify recurring patterns and predict potential issues before they occur. This allows operators to intervene proactively, optimizing resource use and ensuring a more efficient and sustainable production process.
Results
Thanks to the implementation of FactorAI, the results are significant:
- A 10% reduction in energy costs thanks to the analysis of waste and the optimization of machine parameters;
- A 8% reduction in machine downtime through predictive analysis and timely operator intervention;
- A 20% reduction in waste from out-of-spec productions, improving product quality and reducing wasted raw materials.
Conclusions
In the textile sector, FactorAI can bring significant added value, with limited integration time and costs, regardless of the variety and complexity of data sources.
Another key differentiator is the custom indicators built based on the company’s specific requirements. By moving away from standardized reports that add little to the precise data, these indicators have enabled top management, along with the operational insights generated by predictive models, to pursue the goal of greater corporate sustainability effectively.
