Engineering & Mining Journal

MAY 2019

Engineering and Mining Journal - Whether the market is copper, gold, nickel, iron ore, lead/zinc, PGM, diamonds or other commodities, E&MJ takes the lead in projecting trends, following development and reporting on the most efficient operating pr

Issue link:

Contents of this Issue


Page 36 of 83

MAINTENANCE MAY 2019 • E&MJ 35 can cut the time required for planning by 20% to 50%, increase equipment uptime by 10% to 20%, and reduce overall main- tenance costs by 5% to 10%. Implementing it usually requires a multi-step process that begins with es- tablishing an operational data infrastruc- ture — such as OSIsoft's flagship open- enterprise data collection platform, called PI System — to capture data, followed by enhancing and conceptualizing the data. In other words, giving it meaningful con- text. The third step, implementing condi- tion-based monitoring, serves to identify the conditions that lead to an eventual failure of components on an important as- set, and prepares an organization for full PdM 4.0 — the ability to apply analytics and pattern recognition tools to provide real-time, actionable intelligence to auto- matically determine patterns that lead to an eventual machine failure. PI System has been used by several pro- ducers, including Syncrude, a major player in the Alberta oil sands industry. In an ex- ample provided by OSIsoft, Syncrude want- ed to apply event synthesis on their mining equipment fleet for early intervention of maintenance problems and to reduce costs. The fleet also includes a large number haul trucks as well as other production and sup- port equipment such as shovels, graders and dozers. Manual analysis of truck sensor data sets proved too cumbersome for timely analysis and intervention. By using OSIsoft's PI System to create a solution for reporting mechanical events occurring on the equipment, they were able to optimize and streamline calcula- tions and integrate these with notification systems as well as validate and tune per- formance. This involved collecting data from 6,600 data points on 131 heavy-du- ty trucks and five shovels. The results, according to OSIsoft, were impressive. Syncrude calculated that fleet operating expense savings came to $16.75/hour per unit, which equates to a $20 million annual operating cost avoidance, not in- cluding probable lost production hours. AI Gains Ground Over the past year or so, mention of AI has crept into almost every crevice of the mining technology landscape, and main- tenance is no exception. Producers are turning to AI for deeper insight into big data in order to recognize trends and de- termine decision points. In January, Vale SA inaugurated an Ar- tificial Intelligence Center at Tubarão in Vitória, Brazil, that will serve all of Vale's operations around the world. The compa- ny said teams connected to the AI Center were working on 13 projects jointly with the company's ferrous, base metals and coal business areas. The primary focus was on optimizing maintenance of assets such as its haulage trucks and railroad facilities, along with improving manage- ment of ore processing and pelletizing plant processes, improving environmen- tal controls, health and safety prevention and corporate integrity enhancements. Teck Resources has used sensors and data to monitor the health of haul trucks and manage repairs and preventive mainte- nance since 2011. It's now using machine learning — a branch of AI — to take anoth- er step forward, through a partnership with Google Cloud and Pythian, an IT products and services company. Teck said it is "… unlocking new insights from the millions of data points generated by our mobile fleets. Issues that were previously unpredictable, such as potential electric failures, are now being identified before they happen by ma- chine learning algorithms. We are also mod- elling and predicting remaining life span of our trucks, determining wear and wear, identifying abnormal failures and enhanc- ing alarm and notification systems." Meanwhile, AI technology special- ist Uptake and Chilean copper producer Codelco are working together to support Codelco's digital transformation. Uptake said Codelco will deploy AI to monitor the health of mining equipment to anticipate maintenance needs. The current agreement involves min- ing and processing equipment at Codel- co's Division Ministro Hales (DMH) mine in Calama, Chile, including haul trucks, grinding mills, roasters, crushers, pumps, among other equipment with a view to creating an enterprise-wide Asset Perfor- mance Management solution across all Codelco operating mines. Uptake, which a few years ago helped Caterpillar develop a digital analytics platform, said its APM software solution improves operational efficiency by lever- aging AI to create value from operational data. Its flagship product, Uptake APM, builds on what was formerly known as As- set Perform, a product used widely across major industrial sectors. The company said Uptake APM inte- grates key features of what was formerly Asset Performance Technologies' Preven- tance maintenance solution, including the Asset Strategy Library (ASL) — tout- ed as the world's most comprehensive database of industrial content including equipment types, failure mechanisms and maintenance tasks. Uptake acquired the ASL through its 2018 acquisition of Asset Performance Technologies. What's Ahead? It's clear that emerging technologies and strategies such as those listed above as well as Augmented and Virtual Reality, 3D print- ing of components, and vendor-managed parts inventory, to name just a few of many promising concepts waiting in the wings, offer enormous potential to boost main- tenance productivity in the coming years. Most of those benefits will ride on the back of increased digitization initiatives that, in turn, require reliance on sophisticated sen- sors, faster data communications and high- er computing power. The question is, will the ultimate outcome validate the oft-re- peated promise of high-tech evolution: to uncomplicate workers' jobs in an increas- ingly complex industrial environment? Main elements of OSIsoft's PI System, an open-enterprise data-collection platform.

Articles in this issue

Links on this page

Archives of this issue

view archives of Engineering & Mining Journal - MAY 2019