White Paper

How Data-Driven Insights Are Vital For Profitable Edible Oil Production

How Data-Driven Insights Are Vital For Profitable Edible Oil Production

Producers of edible oils and related by-products are best served by designing new processing plants and upgrading existing ones with highly integrated, data-driven digitalization technologies, which can help lower fixed costs to boost operating margins.

A new feedstock for producers of edible oils: deep data insights

Edible oil and related by-product production is capital intensive, requiring large facilities to provide economies of scale. With those facilities come enormous fixed costs such as labor, energy, maintenance, regulatory compliance, and depreciation. But the big margin killers for producers can be the variable costs associated with fluctuating prices of such inputs as soybeans, oilseeds, and corn.

Of course, the industry has long used futures markets to help manage this variability and, when it can, pass on higher costs to customers, wholesale and retail. Now, however, they have new tools to help boost profitability by lowering fixed costs in the form of advanced process and control technologies.  

More and more, industry producers are using these technologies to achieve end-to-end digitalization of their operations. This modernization is also helping them to maximize their production throughputs and the utilization of their largest capital assets — their processing facilities. Today, both the business and competitive imperatives for end-to-end digitalization have become so compelling that the industry should consider adding a new, different type of feedstock to its operating models: deep data insights.

Time to get aboard. After all, in industry after industry, from automobiles to aerospace, operating models driven by real-time data are delivering a decisive competitive edge to those companies using them. For operators in the edible oils industry, not only can this help them prevail over traditional market rivals but also over upstart, lower-cost producers of edible oils and by-products from alternative sources. 

This paper aims to help readers understand the potential scope of digitalizing their operations as well as the extraordinary benefits of doing so. It also offers a four-step approach to planning and executing this transformation. Many of today’s oil processing plants were built decades ago when process control and automation technologies were nowhere near as sophisticated as they are today. Newer plants that were built with lowest cost as their main design criteria may also be missing out on the opportunity these technologies can offer. Either way, this paper can help decision makers consider their upgrade options and how to best implement them to ensure the lowest total cost of ownership.  

How digitalization can drive production efficiency, flexibility, and visibility

The complete digitalization of an edible oils processing plant can lead to dramatic increases in operating efficiency, flexibility, and visibility. While new efficiencies will obviously help cut costs, flexibility can enable those operators who must frequently switch from different inputs to do so more quickly, such as specialty fats producers. And greater end-to-end visibility of operations can assist producers in keeping watch on quality control, meeting their many regulatory mandates, and conducting their track-and-trace genealogy procedures with less labor and fewer errors.

Edible oil plants have many common characteristics. For example, most production steps are continuous processes, such as conditioning, pressing, extraction, degumming, neutralization,  and deodorizing, while others operate in batch mode, such as bleaching, transesterification, and hardening. Product-specific parameters control the various production process cells, which operate sequentially. And storage tanks are often central, shared equipment.

These production plants generate a wide range of control and data-gathering requirements. Among them is the need to:

  • Control multiple sets of process parameters
  • Handle simple sequences of processing jobs on the process cell level
  • Know where materials are located in the plant
  • Keep track of storage conditions and times
  • Control transfers between tanks
  • Track the production history of each final product
  • Control and document product quality data
  • Archive production data and provide reports.

Examples of digitalization trends. Across the edible oils industry, three trends are emerging as representative examples of how advanced automation and control technologies can enhance production performance:

  • Optimizing production flows. Increasingly, plant operators are installing in-line sensors along with automated controls to help improve the speed and precision of their process control. Previously, they might have had to wait as long as 12 hours for the results of a lab analysis before they could use that information to adjust their process settings — in effect, “flying blind” in the meantime. Today, automated controls can use real-time analytics of data streams from in-line sensors to react immediately to process fluctuations, helping achieve a better-quality product at a higher yield. One example is the setting of the separators in an oil refinery: Continuous monitoring of the oil losses in the soapstock and the gums can help operators optimize the settings of the separators.
  • Improving quality control. Real-time analysis of both process and quality data can alert operators to the production of unwanted compounds, such as 3-MCPD (either 3-monochloropropane-1,2-diol or 3-chloropropane-1,2-diol), the most common member of suspected cancer-causing, chemical contaminants called chloropropanols. Compounds like these that exceed acceptable limits can spoil production runs, costing a plant time and money, while also jeopardizing delivery schedules. Again, the combination of in-line sensors communicating in real time with higher-level programmable logic controllers (PLCs) can automate effective responses to the formation of unwanted compounds, reducing the risk of ruined production runs.
  • Boosting operating visibility. If used at all, early generations of sensors and PLCs found in many production plants have limited functions and data outputs. Because plant machinery often lacks networking, readings are often taken manually, to be either relayed to the lab or otherwise entered into higher-level systems by hand. Manual data collection is not only laborious and time-consuming, but it also can introduce latencies and errors into the production process. Industrial networking, using global standards like ProfiNET industrial Ethernet, can eliminate these issues by interconnecting field-level sensors with PLCs and those with higher-level control systems like manufacturing execution systems (MESs) and enterprise resource planning systems (ERPs). These, in turn, can provide greater plant-wide operating visibility and, by extending them, enterprise-wide visibility across all of a company’s production facilities. Company managers can then establish key performance indicators (KPIs) and operating dashboards that enable them to quickly check the operating health of their businesses.

Imagining a fully digitalized edible oils plant

What does a fully digitalized edible oils plant look like? In Figure 1, the five levels of a plant are illustrated to show the real-time integration of data from the lowest field level of sensors, valves, switches, and other purpose-built devices all the way “northbound” to the highest enterprise levels just described in the previous section.

Integrated operations can connect the data spawned by all of an edible oils plant’s operations, from feedstock (oil seeds such as soy, canola or sunflower) arrival and storage through end-to-end processes to storage and shipment of final outputs. For example, a rugged Siemens SITRANS LR460 radar level transmitter with a 24 GHz directional beam can be installed near the top of each feedstock storage silo.

This “sensing fabric” will feed its data to the PLCs, that are programmed to manage both the data and devices at the control level, while also communicating all operational data in real time over industrial Ethernet and highly secure wireless networks, whether WiMAX, cellular, or satellite, to even higher levels.

Ultimately the data can arrive at an enterprise-level cloud that’s accessible from anywhere in the world by almost any smartphone, tablet, or laptop with a network connection.

Bye-bye clipboards. With digitalization, a plant no longer needs technicians with clipboards in their hands, making time-consuming rounds of various plant machines to collect data while also ensuring their sound operation.

Instead, data is collected and transmitted in real time automatically, with PLCs responding immediately to any pre-set parameters to open or close valves, start or end a process, raise or lower temperatures, or manage any other process activity. Compliance and reporting for regulatory requirements becomes much easier, as does implementing key industry standards, such as ISO 9001:2000, and contains Hazard Analysis and Critical Control Points (HACCP), a systematic preventive approach to food safety.

At the same time, the sensors and PLCs can provide self-diagnostics and predictive maintenance that can enable self-healing systems whenever possible. If human intervention is needed, an alarm is issued and recorded, dispatching a technician with precise information about the work order along with any parts required. While hardware functionality will remain important, the key to all these capabilities is software.

The power and capabilities of today’s PLCs go far beyond the microprocessors in distributed control systems (DSCs) and early PLCs deployed several decades ago. What’s more, huge improvements in software and firmware engineering have helped to harness all this computing power into the automation of just about every production process known. 

Four steps to a data-driven operating model

All plant products and machines create data during operation. By collecting, storing, and analyzing this data in a comprehensive, integrated way, plant operators can gain additional systemic benefits over and above the individual benefits that a single, “data-aware” smart component might be able to provide.

Consider, for example, a centrifugal pump motor that is connected to both a smart motor starter and process instruments that measure capacity and the pressure of the flow through the centrifugal pump. On the one hand, the process instrument data can be analyzed to manage optimization of a process. But on the other hand, that data can also be analyzed to assess the pump’s operating condition and overall health, so periodic and preventive maintenance can be planned.

In other words, an integrated system with motor, motor starter, and process instruments can have the additional benefit of a condition monitoring system besides the data benefits of the process instruments. This condition-monitoring concept can also be applied to the following plant operations:

  • Energy Monitoring
  • Tracking and Tracing
  • Process Management
  • Tank Management
  • Power Control

Thinking in terms of a plant life cycle. This concept can be extrapolated to the complete life cycle of an edible oils plant and to the complete supply chain of its owner company’s enterprise.

The challenge when designing a plant is to identify data requirements up front and use the expected use and benefits of the data to drive plant design. In the past, plants were designed with obvious efficiencies and lowest operating costs in mind, but designing them with data sources and flows was typically an afterthought, if thought about at all. Today, the availability of advanced, data-driven process automation and control technologies make the former approach out of date.

The costs to tap this potential in the planning and design phases of a plant are low compared to the modification costs a few years later in an existing plant. While full digitalization of a plant’s operations may be out of reach, given available resources of time and capital, the goal would be to at least create  awareness in the planning phase of digitalization’s potential, and then agree on a strategy to implement the most cost-effective solution. A vendor such as Siemens, which carries a complete portfolio of integrated hardware and software products, can be a good source for learning how to implement digitalization and realize its benefits.

Based on Siemens’ decades of experience in automating millions of factory production processes worldwide, here are four key automation design principles for plant operators to consider if they want to accelerate their migration toward greater automation:

  1. Eliminate. Map processes in detail, then get rid of all unnecessary steps. Advancements in process automation technologies can render many process steps superfluous. If a step doesn’t add value, cut it out.
  2. Simplify. After eliminating as many steps as possible, combine those remaining wherever feasible. Also think about simplifying the context of a process. That is, identify and eliminate (or minimize) external physical, data or schedule I/O dependencies. Reduce or eliminate custom engineering, which always adds cost, time, and maintenance issues.
  3. Standardize. Design and implement digitalization projects, using open standards and uniform interfaces as much as possible, to enable the use of lower-cost commodity hardware components. Standardized solutions can interoperate with legacy systems as well as those from other vendors. They also make installations and reconfigurations much easier and scale better. Maintenance, repairs, and keeping spares become more economical, too.
  4. Virtualize. Move as many hardware-based functions like relays, switches, and terminals to software, which can then be reprogrammed as needs change. Software code can be stored in libraries and re-used across many different deployments. Wireless and cloud technologies can offer further cost savings because expensive infrastructure like cabling and data centers can be eliminated or greatly reduced, the latter by leasing shared cloud-based facilities instead of building and operating your own.

Conclusion: Best time to start digitalization is now

Plant operators across the edible oils industry could realize big benefits by migrating their automation and control of operations from outdated, inefficient, and costly legacy systems to modern, digitalized ones. But the risks of operational disruptions and downtime can cause concerns that might prevent the needed decisions and actions to move forward. 

To address these issues, plant operators should consider careful migration strategies, such as those just described. And they should keep their eyes on the many capabilities and benefits of migrating their legacy automation and control systems to the much more powerful capabilities of today’s PLC technologies. These include:

  • Open yet common architecture for plug-and-play interoperability
  • Framework engineering
  • Real-time and remote diagnostics
  • Highly scalable, high-speed communications
  • Multilayered security

Boosting profitability and competitiveness. Separately or together, these capabilities can help producers make their plant operations much more efficient, flexible, and visible. By doing so, they will also make them more profitable and competitive. Instead of asking if they should migrate their legacy systems, operators should ask how to best upgrade and when to do it.

Experienced assistance coupled with good planning can minimize if not eliminate production disruptions, and stepwise upgrades can often be done more quickly than might be imagined. With a digitalized automation and control system in place, an edible oils production facility can potentially pay for itself quickly. It’s time for plant operators to act, so their decisions to upgrade today can start paying off sooner rather than later. Eventually deep data insights will no longer be a new feedstock for edible oils production but a required one, so early adopters will be the ones to gain the greatest competitive edge from the digitalization of their plants.