Automation and digitalisation

Digital technologies are revolutionising – or at the very least influencing – nearly all aspects of our lives. By no means an exception, the dairy industry is undergoing a digital transformation in which as much value is placed on information as on critical product parameters like solids, fat, or protein. Regular access to accurate production data can facilitate more effective daily and long-term decision-making, allowing for continuous improvements in profitability, productivity and quality. When properly applied, automation and digitalisation enable improved operational efficiency, error prevention, and more integrated processes for streamlining production.

Operational efficiency

Automation and digitalisation do not simply mean replacing manual tasks with computers and robotics. They have concrete, measurable benefits such as streamlining processes, minimising waste, and maximising throughput. Some modern dairies have fully automated lines that adjust parameters in real time, ensuring optimal energy usage and consistent product quality. It is also possible to use automation to drive an inventory management system that efficiently replenishes ingredients in order to minimise downtime and reduce costs.

Quality control

Digital tools enable more consistent and stringent standards of quality control. From inline sensors that monitor milk composition to AI-driven defect detection in terms of packaging, digitalisation facilitates more precise quality control than ever before. For instance, smart control systems can catch subtle product variations, preventing off-spec batches from leaving the plant. When deviations occur, digital traceability allows swift identification and corrective action.

Nurturing the workforce

While many benefits are associated with recent advancements in digitalisation and automated systems, a skilled workforce is still the most central component of any dairy operation. Digital solutions can help companies enhance worker engagement and retention. For example, intuitive mobile app empowers operators with real-time insights, fostering a sense of ownership, or predictive maintenance alerts can prevent unexpected breakdowns, reducing stress and boosting morale.

The digital dairy

Many producers are digitalising their operations to some extent, whether using a few digital tools for specific processes or fully automating their plant’s operations. Starting from a secure foundation of core automation and building digital capabilities over time is becoming the conventional approach in modern dairy processing. The most effective automation and digital systems use domain knowledge, providing producers with insights, allowing improvements in different business dimensions as food safety and quality, sustainability, efficiency, logistics or consumer engagement.

Automation and digitalisation main concepts

In dairy manufacturing, automation can play a crucial role in safeguarding food safety and improving operational efficiency. The set of hardware and software components deployed in automating a manufacturing facility is known as an industrial control system (ICS). These systems are typically organised into layers or levels, each one responsible for specific tasks:

• Layer 1 – field functions
  On the plant floor, sensors monitor machine and process conditions, such as temperature, pressure, and flow. In parallel, actuators, including motors, pumps, and valves, execute commands coming from the process control system (layer 2). This level of interaction is where process data collection begins.
• Layer 2 – automation functions
  • Machine/process control 
    Programmable Logic Controllers (PLCs) execute control logic based on input signals from sensors and other devices. They process this information and determine the appropriate actions to take. Also, ensure that processes remain within specified limits. If a parameter deviates from the setpoint, the PLC can trigger corrective actions and alarms. PLCs play a critical role in safety systems. They monitor emergency stop buttons, safety interlocks, and other safety devices.
  • Supervisory control and data acquisition (SCADA)
    SCADA systems offer real-time local and/or centralised monitoring and control. Through graphical displays (HMI screens), operators can visualise process variables, alarms, trends, and historical data. They allow the implementation of security measures to protect against unauthorised access. Role-based access control restricts operator privileges, ensuring only authorised personnel can modify settings. To enhance reliability, SCADA can also employ redundant servers and communication paths. If one component fails, the system seamlessly switches to backups.
• Layer 3 – manufacturing operations and management functions (MES & MOM)
  MES (manufacturing execution systems) and MOM (manufacturing operations management) systems bridge the gap between enterprise-level systems, such as ERPs (enterprise resource planning) and the plant floor. Here are some of their key capabilities:
  • Real-time monitoring provides real-time visibility into production processes, allowing operators to track equipment status, material flow, and work orders.
  • Production scheduling optimises production schedules, ensuring efficient resource utilisation and timely execution of orders.
  • Quality management monitors quality parameters, performs inspections, and manages deviations to maintain product quality.
  • Inventory control tracks raw materials, work-in-progress, and finished goods, updating inventory levels as production progresses.
  • Traceability enables product traceability by recording data related to batches, materials, and production steps.
  • Workforce management assigns tasks and manages workforce training and cooperation.
  • Maintenance and downtime tracking schedule maintenance, log downtime events, and support predictive maintenance.
  • Integration – integrating an operation’s systems, such as PLCs, SCADA, ERP, quality management systems and warehouse management systems, helps ensure a more seamless data exchange across the manufacturing process.
• Layer 4 - enterprise resource planning (ERP)
  ERP systems manage various business activities, including accounting, project management, procurement, and planning. These systems can be integrated with manufacturing execution systems (MES), allowing production planning to be executed seamlessly. Additionally, inventory updates from MES to ERP occur immediately after materials are consumed and end products become available. Together, PLCs and SCADAs form the backbone of industrial automation. They must be well designed and implemented, securing a solid foundation where food safety and product quality are assured and losses are minimised. On the other hand, MES and MOM systems enable the optimisation of production processes for efficient manufacturing. They enhance productivity, reduce costs, improve quality, ensure compliance, and provide a competitive edge by seamlessly integrating data across the entire production cycle. With the rise of smart sensors, IoT gateways and edge devices, it is possible to bypass traditional stack-and-create applications without needing data from PLC/SCADA systems. Some examples are solutions like asset condition monitoring and performance management.
• The integration of automation and digitalisation tools into production processes enables manufacturing digitalisation. This transformation, which leverages data generated within the plant, is being boosted by advanced technologies such as data analytics, machine learning (ML), and artificial intelligence (AI), leading to enhanced efficiency, cost savings, rigorous quality standards, customisation capabilities, and a more engaged workforce.

Automation and digitalisation – how it works

Control methods

Dairy manufacturing employs various control methods and sometimes a combination of these methods exists within a single production facility. Let us explore each of them:

Manual control

Operations are performed manually, without any automatic sequencing or interlocking functionality. Valves are operated manually. Motors and pumps are started or stopped using push buttons on panels. While certain single valves, such as the diversion valve in a manual pasteurizer, may have automatic control, the overall unit or equipment is still classified as manual.

Unit/equipment control

For a single unit, like a pasteurizer or filling machine, for example, full process control is achieved using PLC (programmable logic controller) and SCADA systems, at minimum. The unit operates from its dedicated operator panel. Each unit follows standardised communication protocols with other units and supervisory systems. Communication occurs either through a limited number of I/O signals or via a communication link. Due to the system’s low complexity, the demands on the local service organisation are minimal.

Line or plant control

Full process control with PLC(s) and SCADA is in place for a complete line, process area or entire site. Operators supervise the plant or line from one or more User Interfaces. Process units, each equipped with its own operator panel, are normally also supervised from central user interfaces. Coordination of routings and operation of units is done from one or more plant PLCs.

Effective line or plant control and supervision provide a comprehensive overview of the facility, enhancing plant functionality. Operations can be sequenced, minimising losses through interlocking functions. Improved communication between units and the plant’s PLC is essential. Additionally, higher functionality is demanded from network and IT components, placing greater requirements on the local service organisation.

Production management – more autonomous operations

Production and cleaning activities are organised into jobs or batches, utilising recipes. Typically, these processes are connected to scheduled production orders that are integrated with the ERP system. The production manager can schedule batches from an operator station, which might be situated in an office or control room. The process operator supervises the execution of planned batches from one or more operator stations.

PLCs manage the routing of batches within a particular line, while a plant server coordinates all activities within the facility. Production history is stored in a database, and the sequencing of batches or operations within a batch can be optimised to reduce product losses or use of other resources. Plant performance analysis allows for optimising operations on the long term. Recording production history also allow for traceability of products. Due to the complexity of sophisticated control systems, it will be necessary to make adjustments according to different plant models, recipes, and programs when process changes occur.

Except for manual control, the other models allow vertical integration, where MES systems can be deployed and integrated into ERP or other manufacturing systems.

What should be expected from a control system?

An industrial control system (ICS) for dairy production needs to have a certain set of capabilities. Functional requirements define the system’s intended behaviour, such as process control, quality assurance and traceability. Non-functional requirements, on the other hand, specify characteristics like reliability, security, and performance.

Here are some key functional requirements for dairy operations:

1. Process control

Manage and control the physical processes. It ensures consistent operation, monitors variables (such as temperature, pressure, and flow), and takes control actions (activates valves, and pumps, and adjusts control parameters) to reach desired outcomes.

2. Quality assurance

Ensures that every batch of dairy products meets high standards of excellence, food safety, and consistency. It builds consumer trust and protects brand reputation.

3. Traceability and compliance

Traceability ensures products can be swiftly traced from gate to gate, enhancing accountability and safety. Compliance with stringent regulations guarantees adherence to quality standards and mitigates risks of penalties and legal liabilities.

4. Batch and production orders management

Ensures efficient handling of production batches. It involves scheduling, tracking, and coordinating processes to optimise resource utilisation and maintain product consistency.

5. Inventory management

Involves tracking raw materials, finished products, and supplies. It ensures efficient stock levels, minimises waste, and supports timely production and distribution.

6. Energy efficiency

Implies optimising energy usage for processes like refrigeration, pasteurization, and packaging. It aims to reduce costs and minimise environmental impact.

7. Asset management

Involves efficiently tracking and maintaining physical assets such as components and machinery. It ensures optimal utilisation, improved maintenance, and cost-effective operations.

In addition, non-functional requirements (NFRs) are crucial in shaping an effective ICS. These attributes focus on how well the system operates rather than its specific functionality. Here are key NFRs for an ICS:

1. Security

• Ensuring protection against unauthorised access, data breaches, and cyber threats.
• Implementing robust authentication, encryption, and access controls

2. Reliability

• The system's ability to perform consistently and predictably
• Minimising downtime and ensuring continuous operation

3. Performance

• How fast the system responds to requests
• Efficient execution of control tasks

4. Maintainability

• Ease of updating, modifying, and maintaining the system
• Clear documentation and well-structured code

5. Scalability

• Handling an increase in users, devices, or workload
• Adapting to changing demands without compromising performance

6. Connectivity

• Refers to the ability of devices, sensors, and controllers to communicate with each other and external networks
• Enables real-time monitoring of process critical parameters and events, base for operational insights, enhancing decision-making
• Facilitates remote access connections allowing technicians to manage ICS equipment from anywhere, reducing travel costs and maximising efficiency

7. Usability

• The system should be easy to learn for both novices and experienced users
• It must be efficient for frequent users
• Users must understand what the system does
• Users should feel satisfied with their interactions

Cybersecurity – building resilience

In an era of rapid technological advancement, the manufacturing industry is undergoing a transformative shift toward Industry 4.0 - a convergence of digital technologies, automation, and data-driven processes. While this revolution promises increased efficiency, productivity, and connectivity, it also introduces new vulnerabilities. Cyber threats targeting manufacturing environments have escalated, posing risks to IT/OT infrastructures, intellectual property, and supply chains. Here are some key points to take into consideration:

1. Operational resilience and business continuity

• Robust cybersecurity measures are essential for maintaining operational resilience.
• Protecting against cyber incidents ensures uninterrupted production and business continuity.

2. Costly downtime and financial losses

• Cyber incidents effecting manufacturing systems can lead to costly downtime.
• Reputational damage, regulatory issues, and financial losses are potential consequences.

3. Mitigating risks

• Investing in cybersecurity measures is crucial.
• Network segmentation, encryption, intrusion detection systems, and employee training help mitigate risks.

In summary, safeguarding manufacturing systems from unauthorised access, manipulation, and disruption is critical for smooth operations and product quality.

Digital transformation

Digital transformation in manufacturing involves integrating technology to improve processes, productivity, and decision-making. However, it’s crucial to consider the human factor throughout this transitional process. Here are some considerations to be taken into account:

• Empowered decision-making: technology applications should empower employees to make more informed decisions. Real-time data, analytics, and AI can enable better choices, leading to improved efficiency and agility.
• Upskilling and collaboration: digital transformation can provide opportunities for upskilling and cross-functional collaboration. It can help employees learn new skills, adapt to changing roles, and work more seamlessly across departments.
• Talent attraction and retention: a digitally transformed workplace may appeal to a wider or more specialised pool of prospective co-workers. Modern tools and a tech-savvy environment can also help enhance job satisfaction and improve retention rates.
• Workplace safety: implementing technology can enhance safety protocols. Predictive maintenance, IoT sensors, and AI-driven risk assessments can contribute to a safer work environment.
• Successful digital transformation in manufacturing requires a balance between technology adoption and human-centric considerations. Companies must prioritise both to achieve sustainable growth and competitive advantage.

Here are some recommended steps towards a digital factory:

1. Prioritise

Begin by identifying process challenges, areas of losses or blind spots. These are the precise areas where investments in digitalisation will lead to significant returns.

2. Automate or update

If the prioritised area relies on manual operation, automation becomes essential. Conversely, if it is an outdated automation system where connectivity is either impossible or excessively complex and costly, upgrading to a modern automation solution is advisable.

3. Connect

After confirming that information can be retrieved from the prioritised equipment or process area, the next step is to choose a scalable and expandable connectivity method and technology, with a proven footprint. This ensures that your investment remains secure as your operations grow. Finally, select the relevant data to share and connect it to the digital infrastructure.

4. Visualise

In the scope of data management, achieving transparency is paramount. The process begins with aggregating and contextualising data, followed by the creation of user-oriented reports and interactive dashboards. These visual representations empower the target audience to gain meaningful insights tailored to their individual requirements. By providing clear visibility into complex datasets, organisations can make informed decisions and drive impactful outcomes.

5. Optimise

This is where the true value of digitalisation becomes evident. Once data is available through visualisation tools, organisations can then optimise processes and tasks through various strategies:

Identifying patterns and anomalies

• Stakeholders can quickly spot trends, outliers, and irregularities. So, identify areas for improvement or potential risks.

Performance monitoring

• Dashboards and real-time visualisations provide continuous insights into performance metrics. Teams can monitor key indicators, such as efficiency rates, quality scores, or inventory levels, and take timely actions based on the visual feedback.

Root cause analysis

• When issues arise, reports and visual representations help trace the root causes. Teams can drill down into specific data points, examine correlations and determine where the deviations occurred.

Process optimisation

• Workflow visualisations reveal process bottlenecks or redundant steps. By streamlining workflows, organisations can reduce lead times, enhance productivity, and minimise waste.

Collaboration and decision-making

• Visualisations facilitate cross-functional collaboration. Teams can discuss data insights, align on strategies, and make informed decisions collectively.

Effective optimisation relies not only on visualisations but also on interpretations and actions based on the gained insights.

6. Predict and auto-adapt

With the integration of edge technologies like machine learning and AI, manufacturers gain the ability to anticipate trends and potential failures. By analysing historical data alongside real-time information, these systems can proactively identify patterns and optimise processes. Predictive analytics tools find valuable applications in manufacturing. Here are some use cases:

Quality control and defect prediction and detection

• Predictive models analyse historical data to identify patterns related to defects or quality issues. By predicting where and when defects are likely to occur, manufacturers can take proactive measures to prevent them. AI-powered vision systems excel at detecting defects that may elude human inspection or traditional methods.

Process optimisation

• Predictive analytics analyses production processes in real-time, considering factors like ingredient characteristics, quality parameters behaviour, and process critical variables. It then suggests real-time adjustments to control variables, optimising process yields and enhancing product quality.

Maintenance optimisation

• Predictive maintenance tools use sensor data and machine learning to anticipate equipment failures. By predicting maintenance needs, manufacturers can reduce downtime, extend equipment lifespan, and optimise maintenance schedules.

Energy efficiency

• Predictive models analyse energy consumption patterns. Manufacturers can adjust production schedules, equipment usage, and energy sources to minimise costs and environmental impact.

These tools are designed to enhance decision-making by identifying and mitigating problems in advance to improve efficiency and productivity in manufacturing.

Final reflections

Within the dairy processing industry, the adoption of automation and digitalisation has already transformed production facilities and will continue to do so. Leveraging technologies like the Internet of Things (IoT) and artificial intelligence (AI), dairy processing plants now benefit from tools powered by real-time contextualised data, streamlining processes, and improving efficiencies. Automated production lines ensure consistent quality and reduce human error, while predictive maintenance algorithms optimise equipment uptime. As digitalisation continues to evolve, we can expect even more exciting developments that will shape the future of dairy product manufacturing.