Skip to main content Skip to Footer

BLOG


November 28, 2018
Improving data veracity for metals and mining companies
By: Dr. Andrew Zoryk

As I write this blog, it is the 50th anniversary of one of the most successful and influential rock bands ever—Led Zeppelin. With approximately 300 million records sold, their unique sound led to the creation of the music genre “heavy metal.” While Led Zeppelin reigned as the biggest band in the world in the 1970s, their image was often shrouded in mystery and stories—people were never quite sure what the “real truth” was!

Fast forward to today and the metals and mining industry faces a similar dilemma with “data veracity”—defined as the degree to which data is accurate, precise and trusted. Companies are increasingly relying on a multitude of data sources to operate and transform their businesses; however, many are unsure of the quality of that data—and that’s an enterprise-level existential threat.

These observations are confirmed by Accenture’s Technology Vision 2018 survey, which included executives from metals and mining companies. The survey showed that 82 percent of metals and mining industry respondents are increasingly using data to drive critical and automated decision-making, and 91 percent agree that they are basing their most critical systems and strategies on data. However, only 36 percent of metals and mining respondents perform validation of data and, furthermore, 63 percent are not confident in the reliability of their data.


82% of surveyed metals and mining leaders are increasingly using data for critical and automated decision-making, at scale.  of surveyed metals and mining leaders are increasingly using data for critical and automated decision-making, at scale. 36% of metals and mining executives report that they validate their data and are reasonably to very confident in its quality. of metals and mining respondents report that they validate their data and are reasonably to very confident in its quality.

Source: Metals and mining respondents in the Accenture Technology Vision 2018 survey.


82% of surveyed metals and mining leaders are increasingly using data for critical and automated decision-making, at scale.

of surveyed metals and mining executives agree that their company is increasingly using data for critical and automated decision-making, at scale.

36% of metals and mining executives report that they validate their data and are reasonably to very confident in its quality.

of metals and mining respondents report that they validate their data and are reasonably to very confident in its quality.


Source: Metals and mining respondents in the Accenture Technology Vision 2018 survey.

Proliferation of data sources

Let’s take a closer look at some of these rapidly growing new data sources. At the shop floor level, metals and mining companies are deriving sensor data (e.g., product quality, video, equipment-related) at an exponential rate from a new generation of manufacturing execution systems and advanced process control capabilities. Compounding this growth is the increased resolution of data due to the use of more sophisticated cameras and scans being taken from multiple dimensions.

Outside of the enterprise, abundant new sources of data are coming from the broader business ecosystem. In the past, metals and mining companies typically believed that competitive advantage depended upon keeping critical data (e.g., manufacturing codes of practice, product properties, commercial pricing practices) securely “locked” inside their four walls. While this is still valid, the recent advent of digitalization, increased business-to-business (B2B) collaboration and new business model thinking is changing this perception. Companies now realize that their future competitiveness also depends heavily on many external sources of data.

Case in point: Increased collaboration with upstream and downstream ecosystem partners can improve supply chain and customer-facing agility—for example, through increased B2B electronic data interchange (EDI) integration. Our Technology Vision survey showed that over the next two years, more than 90 percent of metals and mining respondents expect the volume of data exchanged with ecosystem partners to increase. Thirty (30) percent have seen the number of partners they work with double or more compared to two years ago.


93% of surveyed metals and mining executives anticipate that the volume of data exchanged with ecosystem partners will increase. of surveyed metals and mining executives anticipate that the volume of data exchanged with ecosystem partners will increase. 30% of metals and mining respondents say that the number of partners their company works with doubled or more versus two years ago. of metals and mining respondents say that the number of partners their company works with doubled or more versus two years ago.

Source: Metals and mining respondents in the Accenture Technology Vision 2018 survey.


93% of surveyed metals and mining executives anticipate that the volume of data exchanged with ecosystem partners will increase.

of surveyed metals and mining executives anticipate that the volume of data exchanged with ecosystem partners will increase.


30% of surveyed metals and mining leaders are increasingly using data for critical and automated decision-making, at scale.

of metals and mining respondents report that the number of partners their company works with has doubled or more versus two years ago.

Source: Metals and mining respondents in the Accenture Technology Vision 2018 survey.

Given these new sources of data and many others, metals and mining companies are now awash in a “data revolution.” But they’re also grappling with how to make the best use of all this new data and how it will impact their business.

Doubling down on data veracity

As metals and mining business models are increasingly data-driven, data veracity becomes a key consideration. Inaccurate and manipulated information threatens to compromise the insights that companies rely on to plan, operate and grow. Left unchecked, the potential harm from bad data becomes an enterprise-level threat.

As the Technology Vision survey shows, while metals and mining respondents report that they are basing their most critical systems and strategies on data, many have not invested in the capabilities to verify the truth within the data. Clearly this is a dilemma that needs to be addressed.

Some companies have already started this critical work. Let’s look at two examples from the metals and mining industry.

Example 1: Historically, uncertainty about the quality and trustworthiness of mine production data has been a barrier to leveraging its potential. However, the availability today of massively increased computing power is helping to change this.

Look at the Connected Mine. It leverages Internet of Things (IoT) and cloud services to extract large volumes of data in real-time from mining operations (e.g., worker activities, in-pit equipment, GPS) to provide visibility, advanced analytics and predictive capabilities that help improve performance and efficiency across the entire value chain.

By deploying such capabilities, mining companies can improve data veracity through machine learning, change detection and other optimization techniques, which builds greater trust in their data over time.

Example 2: Environmental, health and safety (EHS) compliance has long been an important initiative in the metals and mining industry. One steel producer is using video analytics and machine learning to help minimize unsafe working conditions in a manufacturing plant. The traditional approach for managing safety across the plant involved using a video surveillance system set up in a control room, with approximately 20 screens linked to more than 100 cameras located across the site. Based on Accenture project experience, despite this level of coverage, virtually none of the video results were seen by the control room operators. Furthermore, most safety incidents went unnoticed after 20 minutes of them happening.

To address this issue, a new approach of combining video analytics with artificial intelligence was developed. It used an adapted video source to execute analytical algorithms, as well as an automatic machine learning system to build and train neural networks to detect and decipher patterns and correlations (similar to the learning and reasoning used by humans).

The steel producer’s initial prototypes of this new approach achieved detection reliability levels of more than 90 percent. And data veracity is improved through the learning algorithms, which reduces the effort for rolling out new use cases across the site and increases the company’s trust in data.

Forging ahead with data intelligence

Metals and mining companies can start to address data vulnerability by building confidence in the following three key data-focused tenets:

  1. Provenance, or verifying the history of data from its origin throughout its life cycle.
  2. Context, or considering the circumstances around data use.
  3. Integrity, or securing and maintaining data.

One way to accomplish this is by establishing a “data intelligence” practice, drawing from existing internal data science and cybersecurity capabilities, or outsourcing to a third party with credible experience in these areas. The group can work to embed and enforce data integrity and security throughout the organization, while adapting existing investments in cybersecurity and data science to address data veracity issues.

It’s also important to understand that the presence of bad data in a system is not always the result of malicious intent. Instead, it may be an indication that a business process isn’t working the way it was intended. Using a data intelligence practice to uncover these processes will allow metals and mining companies to “reduce noise” in data so that real threats stand out.

In summary, data is the lifeblood for an increasingly digital metals and mining industry, enabling complex business decisions that can drive sustained growth. Ensuring the veracity of this data, then, becomes a cornerstone of strong performance. Failure to do so can have significant consequences—especially as metals and mining companies invest heavily in autonomous data-driven systems, such as “lights-out” plant operations.

As Led Zeppelin’s 50th year celebration ends, I realize that the band may have provided heavy metal fans an existential experience. But data veracity will give metals and mining companies an unprecedented opportunity to achieve real truth and thrive.

Popular Tags

    More blogs on this topic

      Archive

        SUGGESTED CONTENT