What IoT and the Digital World Mean for the Cement Industry – Steps You Can Take Now
…. A periodic article by Folkert Haag on the cement and concrete industry…. printed in the November, 2015 issue of International Cement Review
Excitement around what has been coined “the Internet of Things” continues to grow as manufacturers, vendors, and customers begin to realize the substantial benefits of embedding sensors in everything from vehicle tires to conveyor belts. The amount of information flowing to and from “smart” products and machines is staggering. According to a report by IDC and EMC , the data being generating is predicted to reach 44 trillion gigabytes by 2020. That means there will actually be more bits of information in our digital universe than there are galaxies in the physical universe!
Of course, data by itself is relatively useless. Companies need to have the intelligent systems and technology backbones in place to capture the data and use it to make smart, insightful decisions. Those that do are seeing substantial return on investment in the form of equipment up-time, reduced costs to ensure quality, optimized logistics, and much more.
Although the cement industry tends to be slower in adopting new technologies, the revenue and margin-impacting benefits experienced by other industries are too compelling to ignore. Some cement executives are under the false impression that bringing their companies into the digital world requires millions of dollars in upfront investment. In reality, the cost of embedded sensors has dropped significantly and learning how to utilize data can be a gradual step-by-step process. Below are few real-world examples of how data is being used to change the business of manufacturing, beginning with a simple application and progressing to the more complex.
Equipment Monitoring – The First Step
By 2020, Gartner analysts predict there will be 526 million pieces of manufacturing equipment capable of communicating through sensor-enabled networks, with 90 million of those designed specifically for the mining industry. In fact, the predicted 25% annual growth rate of connected mining products is among the fastest of the industries tracked by Gartner (WJS, 6/2/2015). These connected machines stream health and status data, which is captured by other machines or by monitoring systems. Should the data fall outside of normal parameters, alerts are triggered, allowing workers to immediately resolve the problem or, in some cases, prompting the equipment to self-adjust. Real-time issue resolution or problem identification can save significant time and money.
Joy Global, an underground mining equipment manufacturer, provides an example of the cost-saving benefits associated with remote equipment monitoring. Installed at a customer’s site, one of Joy Global’s mining machines kept overheating. Engineers thought the power-and-speed adjustment control on one of the motors had failed. But, after reviewing the machine’s sensor data, it was determined a heat exchange device needed replacement – a much smaller repair job. This approach can also be applied to the cement industry. Remote monitoring could be used to oversee operations of big vehicles in the quarries and to report on key metrics such as fuel consumption per ton, tons per shift, or operating hours. It could also be used in crane to crane communication to avoid collisions or to safely move containers.
When analyzing equipment status and performance data to determine trends and even enriching the data set with manufacturing data such as the type of product being produced, it’s possible to predict potential malfunctions and maintenance needs. Using this information, companies can proactively schedule maintenance during times of least impact and avoid unexpected, costly downtime.
Predictive maintenance has significantly lowered costs in the oil and gas industry. For example, when a piece of equipment on a remote oil rig, such as blowout preventers, mud pumps or ship stabilizers, fails unexpectedly, drilling halts until the necessary repair parts can be delivered. This down-time can add millions of dollars to a well’s cost. However, continuously monitoring and analyzing of data transmitted from the equipment, such as temperature, speed and vibrations, is now being used to maintain consistent drilling. This same process can be applied to the operations of cement plants by improving up-time on critical cement equipment such as kilns and grinding mills. Also, sensor data from trucks and excavators can be combined with outside information like usage patterns to predict when a vehicle may fail.
Once a company has implemented remote monitoring and predictive maintenance solutions, it’s possible to take the next step toward predicting product quality. Predictive quality involves reviewing all types of process data such as pressure and electricity, and correlating it to quality parameters. More specifically, sensors track specific characteristics of the product as it is being manufactured, allowing for real-time adjustments to control the resulting quality. One company already using data in this way is Mohawk, a large carpet and rug manufacturer. Mohawk uses sensors to track specific characteristics of the carpet as it is being manufactured, which helps workers make adjustments in real-time to ensure consistency between rolls and to stay on target to meet quality measures. Collecting data also allows for quality certification to reduce costs associated with warranty claims and provides better details of each resulting roll. In addition, the approach enables Mohawk to increase its customer satisfaction by being able to more consistently deliver the exact quality expected by its customers.
The ability to predict quality could be very advantageous in the cement industry as well. High quality cement requires that a very homogeneous meal goes into the kiln. The limestone must be crushed consistently, which requires monitoring the input materials as well as adjusting the speed of the vertical roller mill to gain the best output. Other applications include tracking and measuring the durability of the concrete mixture and reporting on compliance to international standards based on predictive quality of the concrete, instead waiting for analysis results from a laboratory.
Streamlining logistics often affords the biggest opportunity for improvement and the cement industry is no exception. Collecting, comparing and integrating data from multiple sources to identify best practices or anomalies is proving to be useful in making smart decisions that improve production yields and fleet management.
Similar to the way the Internet of Things connects smart devices with farm suppliers and service providers for more precise farming, connected logistics can also improve quarry production. Technology integrates key production parameters such as weather, workforce status and soil conditions, to help management adjust operations. Increasing the yield per quarry on a large scale could have a significant, positive impact to bottom-line profitability.
Fleet management is another opportunity to leverage data insights for increased efficiency. Connected fleets combine vehicle telematics data (tire pressure, engine speed, etc.); driving behavior data (speed, accelerating, braking, time spent loading and unloading); and business data to save on transportation costs. For example, by viewing fuel consumption under various conditions, such as a full load, no load, idling, etc., companies can identify the trips with the highest fuel consumption, drill down to the details, then compare to fleet benchmarks in order to reduce activities leading to higher-than-normal costs.
Driving behavior can be analyzed in a similar way by looking at fuel-consuming behaviors and correlating it to routes or other work information. Finally, vehicle manufactures data is evaluated for key performance indicators such as engine efficiency or maintenance needs. Zoomlion, a manufacturer of equipment for ready-mix concrete production, is actively testing these types of data-based metrics in order to improve their fleet operations.
For sophisticated data-driven organizations, predictive analytics can provide advanced insights into nearly every aspect of the product lifecycle. Using internal and external data, predictive analytics involves finding patterns and building algorithms to forecast future results. Critical information is delivered to the right people, at the right time, so that action can be taken when its needed most.
Joy Global again provides a solid example of the benefits in using predictive analytics to maintain a safe work environment and ensure consistent operations. When machines extract materials from a coal mine, monitoring the integrity of the infrastructure is extremely important. Cavities inside the rock can cause the mine roof to collapse prematurely, stopping production. When this happens, it can take up to a week to clear off the machine and cause significant production loss (WSJ, 6/2/15). However, sensors in Joy Global’s connected machines relay pressure data in the rock above. Combining this data with additional environmental factors, operators can take preventive action by slowing the production rate of the machine, increasing the hydraulic pressure from the roof supports and continuing operations. A similar application could maintain the safety of quarries as well.
The Internet of Things is opening up opportunities we are just beginning to understand. In an environment where every dollar counts, companies must dig into data to uncover opportunities for greater efficiency and effectiveness. Leaders at cement companies willing to apply the lessons learned from other industries and make decisions based on real-time information, will benefit from wider margins, improved financial performance and sustainable success.
Want to learn more?
Attend the SAP Manufacturing Industries Forum 2016 on June 14-15 in Lombard, IL. For details, please visit the forum’s website.
Attend the SAP Forum for Building Materials, July 7-8, Walldorf Germany. For details please visit the forum’s website.
I like the step by step aprroach on IOT.This is a repeating pattern in successful customer projects. Start with vertical integration to gain visibility on productivity and asset availability.
Add root cause analysis, connect operational data from fleet, crushing, mixing and correlate this with the ERP type of information on batches, cost, customers, quality.
Explore machine learning, classification and pattern recognition in the next step - understand the data to a new level, help your engineers to recognize new improvement potential.
Use your learnings and enter into real-time predictions to improve quality, uptime - both contributing to OEE.
While the predictive models and the "cyber" representation of the physical world may be quite different from discrete to process industry, the best practices to explore step by step, often in a co-innovation mode, are very consistent.
So are the value potentials. Just heard this morning from a customer who estimate their improvement potential for OEE at around 4%. How much would 4% better OEE mean for you?