Read today’s article to keep up with Gartner’s recommendations for keeping IoT and manufacturing data trouble-free.
The Internet of Things (IoT) has become one of the most important technologies today as it gathers vast amounts of data on the corporate network, unleashing a wealth of information that will need to be assessed, collected, organized, and analyzed by corporations.
There are already about 7 billion IoT devices connected. However, according to experts, this number is expected to grow to 22 billion by 2025.
Next, you will follow the following topics:
- • Recommendations for addressing IoT challenges;
- • Strategic planning assumption;
- • Impacts and recommendations;
- • Steps of Curating IoT Data Into Insights.
Keep reading the article.
IoT data is transforming manufacturing companies into smart factories as they drive a range of digital business initiatives to optimize processes, drive business in a more targeted way, and, as a result, create new revenue opportunities.
Based on the insights gained, companies are able to improve the operational efficiency of physical assets across the industry, as well as provide more satisfying user experiences.
Companies have been modernizing manufacturing processes, using improved software and IoT technologies, based on the key needs and challenges they face.
By introducing new real-time IoT data, IT staff as well as leaders and directors will intimately understand how manufacturing assets operate, and will also see ways to enhance operations and address any issues that may arise over time.
The “2020 Gartner IoT Implementation Trends Survey,” points out that 26% of companies said they have implemented IoT in large-scale production environments because of the benefits the technology can provide.
However, while there are many benefits provided, there are also challenges that must be considered. The speed and volume of IoT data threaten to overwhelm plants’ infrastructure and business processes, as well as influence staff skills.
The lack of an IoT-enabled Data Lake for Manufacturing data, for example, will cause considerable increases in operational expenses, as well as trigger security vulnerabilities and poor-quality information.
Moreover, the information collected will often not be merged with other manufacturing data and will therefore generate inaccurate and incomplete results, leading to poor decision making and loss of corporate profits.
To prevent problems like these from happening, the Gartner® Report has some recommendations, such as:
Recommendations
CIOs focused on manufacturing digital transformation and innovation who are driving the smart factory initiative should:
- • Implement a data-collection engine that will tag IoT data by contextualizing it into hot, warm or cold data tiers.
- • Establish a data lake that will combine both contextualized hot, warm, and cold IoT data and manufacturing production data per the data schema requirements.
- • Define what type of business insights each critical decision maker requires by auditing the existing dashboards and evaluating the type of data visuals they require.
Strategic Planning Assumption
By 2025, 55% of the global manufacturers will implement an IoT-enabled data lake that will provide business leaders with accurate business insights, up from 25% today.
Impacts and Recommendations
IoT Data Is Not Fully Utilized Because the Volume Being Generated by Physical Assets Makes Generating and Abstracting Impact ful Insights Difficult.
As IoT investments fully materialize within the manufacturing plant, new data streams begin to be collected. These IoT data streams are in abundance by nature and are raw, unfiltered, repetitive and uncurated. All IoT data is telemetry, and the data has a low level of contextualization. When IoT data is first acquired in its raw form, it is immediately tagged as “hot”.
However, as the IoT data gets processed over time, an increased level of contextualization is added with the blending of manufacturing data. As it goes through the contextualization funnel, IoT data will reveal the level of utilization and relevance to each manufacturing decision-maker’s role.
Steps of Curating IoT Data Into Insights
- • Step 1: Acquisiton;
- • Step 2: Contextualization;
- • Step 3: Visualization.
The conceptual view, explains:
- • Acquisition refers to the raw endpoint data coming from the sensors. Raw endpoint data flowing through an IoT architecture is typically high in volume, velocity and variety. At the aggregation point, IoT data is then tagged as “hot.”
- • Contextualization refers to blending of manufacturing data. Contextualization functions can range from simplistic (e.g., filtering) to sophisticated (e.g., classification). Manufacturing data is blended to IoT data, and the data is then tagged as warm (simplistic) or cold (sophisticated).
- • Visualize refers to the descriptive, predictive and prescriptive visualization of the curated manufacturing and IoT data as a by-product of the entire contextualization process.
The model is designed so that hot data has very little contextualization and the insights that are produced from hot data are specific to one use case, while cold data requires manufacturing data and has several business use cases. For example, hot data could be represented as time series telemetry data that can trigger an automatic shut-off if the machine hits its designated threshold of over heating. Cold data would account for down time across several machines across the factory and create insights on why and how often the machines needed to be repaired or maintained to prevent future unscheduled downtime.
Recommendations
- • Organize a “digital transformation” alignment plan that consists of IoT/IT/OTsystems to determine which manufacturing asset to start integrating data from (see Survey Analysis: IT/OT Alignment and Integration).
- • When ingesting IoT data, categorize the hot data into its proper category (for example, telemetry, events, or time series), and tag the contextualized IoT data to either warm or cold as per the data schema requirements.
Conclusion
IoT is a technology that is proving to be increasingly necessary to extract and exploit business value in the manufacturing industry, based on the volume of data.
It is a trend that, according to Gartner, tends to grow exponentially over the next few years. At least 12 million IoT endpoint devices are expected to be purchased every day by 2022.
The concept can especially assist modern manufacturing processes, which produce a huge amount of data.
When collected and used properly, IoT data can provide valuable insights to help make smarter business decisions.
Therefore, it is a good tactic for corporations looking to take a competitive advantage to further leverage their business.
