Skip to main content
Лого на Европейската комисия
Internal Market, Industry, Entrepreneurship and SMEs

FAN-MODE

Quality step-up in FANcy yarns manufacturing by predictive MODEls

Sectors covered
Technology innovation in textile & clothing
Innovation priority
Advanced digitised manufacturing, value chains and business models
FAN-MODE banner

 

Introduction

Pecci is producing yarns for the fashion industry. In recent years, the industry has required more complex yarns, which are more and more difficult to produce. The complexity of these processes has raised several quality issues, which have in turn led to higher costs for Pecci. The status quo is to react to quality issues after they have already become an issue, and the actions are based on the expertise and knowledge of the operators. For these new processes, the company has acquired new types of machinery, this equipment are capable of collecting data related to the production process. However, Pecci is not in a position to give added value to this data in order to find a solution for the quality issues. This is because the operative business does not have the required expertise for advanced data analysis.

The technology provider DatenBerg is focused on the automated analysis of manufacturing data using its specific software. Right now, Datenberg is active in the compounding and chocolate sector. During this project, Pecci will provide access to the issues faced by the textile industry and DatenBerg is delivering the expertise of its data analysis capabilities.

Together, they will implement a demonstrator of automated data-analysis at Pecci’s manufacturing site. This demonstrator will support the operators in finding the root cause of quality issues based on data and by applying prediction models to act proactively before a quality deviation occurs. This will help Pecci with cost-saving, efficiency gains and allow them to support their customer relations within a highly competitive business environment. For DatenBerg, the demonstrator will be a means to introduce themselves in the textile industry market. The demonstrator will be developed in a way that will be easily transferable to other SME’s in the textile industry.

Objectives

Supporting the operator with data-based recommendations at a shop floor level, in particular for the needs of a complex yarn producer. Two use-case scenarios will be developed.

  • Root-cause identification: Support the operator by automatically finding the root causes of quality deviations.
  • Proactive actions: Predicting the quality of the produced part in near real-time based on process parameters.

By implementing this use-case the following objectives below will be achieved.

  • Business objectives: Reduce non-re-workable delivered material to the customer by 30% (equal to 250k EUR per year), and offer an objective data-based support for operators.
  • Technical objectives: Data-analytics in real-time by implementing the DatenBerg software at Pecci’s manufacturing site, automated prediction algorithm (e.g. artificial neural networks, or Bayesian inference) for the textile industry.

Outcome of ELIIT project support

Impact (to date):

  • By the end of January 2021, out of 20 machines at the manufacturing site, 5 have been identified as having easy data access.
  • Amount of data: Spinning and winding machines produce quite a decent amount of data due to the high number of parallel heads, smaller data acquisition issues have been handled jointly and the whole acquisition process has been improved.
  • Analysis: First correlation between machine parameters and the number of failures have been identified, further steps will follow soon.

Partners

SME: Pecci Filati S.p.A.

Country: Italy
Year of creation: 2002
Background: Pecci is a unique yarn producer of all types of yarns for knitwear and handknitting, used by several influential brands in the apparel world.
More information: company website

Technical provider: DatenBerg GmbH

Country: Germany
Year of creation: 2018
Background: DatenBerg is mainly focused on the automotive and rubber compounding sector. It started at the Audi Smart Factory Hackathon and has since entered the manufacturing sector with its machine-learning software.
More information: company website, LinkedIn