Enrico Plazzogna
Enrico Plazzogna, Executive Vice-President, Sales and Marketing, Danieli Automation SpA

The current steel market outlook is characterized by plant underutilization and leading metals producers are seeking to lower capital expenditure, aiming at improving the efficiency of production facilities, upgrading the quality of products, ensuring the health and safety of workers as well as adopting environmentally sustainable solutions. At a time when digitalization is transforming the global steel industry, Danieli has created a new cross-functional business unit named Digi & Met with the mission to develop and implement new plant design concepts, based on digital innovation, and also new business models built on the principles of servitization and outcome economy. In an exclusive interview with Steel 360,Enrico Plazzogna, Executive Vice-President, Sales and Marketing, Danieli Automation SpA, explains the intricate workings of a smart steel factory and the revolutionary potential of data analytics and smart manufacturing to optimize production and make the global steel industry more competitive. Excerpts:

 Enrico Plazzogna

What are the characteristic features of a smart steel factory?

A smart steel factory is operated from a single pulpit, where a few operators supervise the fully automatic production process. Monitors in the pulpit show all the main technical parameters and the reports are generated by a huge amount of data and variables, coming from the automation systems and from smart sensors, highlighting possible quality or process issues and suggesting corrective actions. CCTV camera images are used for plant visual monitoring. Machines are designed to avoid manual operations and robots are used in dangerous areas or wherever operations are repetitive and frequent. Cranes are as much as possible unmanned and supervised from a single pulpit. Condition monitoring systems help in the detection of mechanical anomalies and suggest preventive maintenance. The production programme is optimized in order to process the orders with a schedule which allows the best machine utilization and the lowest OPEX production route, with a focus on final customer satisfaction in terms of quality and on time delivery.

How does smart manufacturing optimize productive volumes?

Many different positive effects derive from smart manufacturing solutions. A completely automated system, including robotics, is more efficient and reduces delays in the production time. Furthermore, optimizing the production schedule with a MES system means improving plant yield. The full knowledge of process parameters and the use of machine learning systems reduce the out-of-quality production, while the knowledge of tools (stands, ladles, gears, rolls, etc.) status avoids unwanted stoppages. A smart plant makes the widest possible use of data analytics and KPIs are constantly monitored to reach the expected targets.


How does Danieli’s Digi & Met platform exemplify the characteristics of industry 4.0?

In Danieli we started working on data analytics and intelligent manufacturing more than 10 years ago, and today we have a dedicated business unit named Digi & Met and a new building where process technologists and automation experts work together to extract the added value that innovative technologies can provide. There is an evident change in the business model, more and more focused on partnership with end-users and on customer support activities, based on the know-how on metals of the Danieli Group. The new approach and the innovative mentality, together with the focus on data analytics, have provided new ideas to improve the products, machines and automation systems that Danieli can offer to the market to allow our customers to be more competitive than other players in the market.

Danieli Industry 4.0
Danieli Industry 4.0

What are the benefits of installing the Manufacturing Execution System (Q3-MET) at an integrated steel plant?

We have measured KPIs in some of our reference plants where Q3-MET was installed and sometimes results exceeded our expectations. An improvement in the yield in the range of 2-3%, depending on plants and configuration, is the consequence of many positive effects, determined by an effective and structured production scheduling. In particular, machine utilization can be improved up to 10%, inventories in the stockyards can be reduced up to 15%. The reduction of shipping time is another important benefit of Q3-MET, as well as the improved control of quality of production. One less measurable but very important advantage deriving from the application of Q3-MET for the entire plant is the unification of operator’s interface and platform to manage production.

How does Danieli’s Q-MELT model enhance the energy efficiency of an EAF plant?

Q-MELT is the perfect example of machine learning system applied to steel industry. In fact the Melt Model, part of the Q-MELT system, automatically changes the set up in the melting process depending on deviation from the expected statistical behaviour of some key variables, based on big data collected during plant operation. The Melt Model is the controller implementing the soft-landing strategy to attain the heat targets (steel temperature and carbon) in the best way. The controller also integrates the estimation of the current bath temperature and carbon, useful to track the bath status towards the end of the process. As the main approach is statistical, the system is constantly learning from the data retrieved from the field, and the control gets more and more accurate over time. The first running plants have produced impressive results, with return on investment in less than 8 months thanks to reduced energy consumption and lower electrode consumption.

What are the relative benefits of Danieli’s condition monitoring system for the steel industry? Is predictive maintenance an integral feature of a smart factory?

These two questions have a common answer. The condition monitoring system is the basis for the application of predictive maintenance.

Most of steel plants rely only on breakdown maintenance, when repair actions are taken only after machine failure, with the consequent downtimes. Preventive maintenance is instead executed on a regular basis, regardless of their condition, therefore not optimizing OPEX for spare parts.

Predictive maintenance uses the Danieli Condition Monitoring System to identify changes in signal values which might indicate a defect which will lead to a failure (typically vibration and temperature). The Danieli Condition Monitoring system is essential for critical parts (gearboxes, motors, shafts, fans, pumps, etc.) in order to avoid unwanted and uncontrolled stops in the plant due to failures. The benefit is the reduction of delays in the plant and, therefore, increased working time, or more productivity, while OPEX is optimized as repairs are done only when needed.

In a smart factory the analysis of variables from DCMS allows for predictive maintenance, contributing also to generating a database of maintenance data, contributing to OPEX reduction and higher reliability of equipment.

Danieli Industry 4.0

How does Danieli ensure quality control prediction?

Danieli Automation has developed an innovative digital solution named Q3-Premium, using breakthrough advances in Big Data Technologies. Huge amounts of data are collected and processed, together with material tracking and feedback from special sensors, in order to have a quality control engine on the entire production, with the capacity to detect potential issues and implement prompt corrective actions. Best practices are continuously improved and consolidated thanks to a data analytics platform providing the statistical framework for optimization and decision support.

In other words, Q3-Premium provides recorded process data with analytical tools for identification of deviations and their influence on final product quality. Once defects are identified, the system applies machine learning tools and predictive analytics models to support the recognition of correlations between quality deviations and process anomalies. In this way a decision-support system is built for implementing reactive and proactive actions in order to fix out-of-quality production and avoid same quality issues in the future.

This is probably one of the most interesting targets of digital technology: an automatic system to improve consistently and continuously final product quality.

How does digital innovation ensure optimal plant utilization and reduction in operational expenditure?

The intelligent plant is a safe, flexible and environmentally friendly concept of manufacturing, where system and equipment autonomously execute complex tasks and support humans in complex decision-making or execute automatic decisions.

In the near future (even already now) flexibility will be the key to success. The capacity to plan quickly and efficiently plant resources depending on market conditions and deciding what to produce depending on profitability of each order are important. Only digitally enabled plants will be able to apply these technologies, only plants with full knowledge of their data, with visibility of KPIs on quality of production, on energy consumption for each plant machine for each specific type of production can apply these technologies. Only plants with a MES system in place are able to re-schedule production according to changes in order precedence.

To sum up, only those plants that have embraced digital innovation will be able to optimize plant utilization and reduce operational expenditure, in order to remain a step ahead of their competitors.