These days, the set of technologies under the “Internet of Things” (IoT) name is considered the enabling factor for a potential Fourth Industrial Revolution. While the hype has been great, the business value may be greater. According to McKinsey, if policymakers and business get it right, linking the physical and digital worlds could generate up to $11.1 trillion a year in economic value by 2025 in different fields (e.g., $3.7 trillion in operations and equipment optimization, $1.7 trillion in public health and transportation; $850 billion in logistics and navigation).
The organizations that use IoT systems are faced with a multitude of opportunities and challenges. IoT also will affect consumers and operators in different ways.
The two main areas of improvement and evolution that have been identified pertain to business models and operations.
One of the main drivers of business value is the integration between operational technologies (OT) and information technologies (IT) that previously were mostly separate and run by different parts of the organization according to their different systems and methods.
Currently, IT and OT are mostly separated by:China’s Digital Transformation
– Different technologies
– Different organizations and responsibilities
– Different skills and professional profiles
– Often even different codes and standards
Segregated OT and IT environments are inherently inefficient and costly. Innovation can stall due to the resulting technological and financial limitations. For example:
– Lack of communication between IT and OT systems prevents enterprises from using control data in business intelligence applications.
– OT organizations cannot take advantage of rapid advances in IT.
– OT organizations cannot take advantage of the cost savings of standards-based solutions.
We can improve automation and accelerate innovation by converging these two worlds in an Industrial IoT (IIoT).
To achieve the potential value and opportunities from the IIoT, integrating OT and IT data is one of the most important challenges for companies, specifically for asset-intensive organizations. Integration of processes and information flows enables them to overcome existing obstacles and obtain some key valuable advantages, such as:
– Critical decisions based on facts rather than approximate information
– No delay between facts, analysis, decisions, and reactions
– Holistic optimization of systems
– Improved decision-making
– Lower operating expenses (OpEx) by minimizing organizational and technological overhead
– Accelerated business results by streamlining development projects
– Reduced risks
To understand how IoT is being deployed by business today, it’s crucial to keep in mind that today’s market is heavily driven by use case scenarios and proofs of concept. Companies need to know which IoT applications have the potential to deliver the most value.
The required ingredients are:
– Data: Detailed time-series data from devices is collected, replicated, or federated, depending on existing technologies; relevant transactional data is made available in the same platform to provide context for analysis; data must be ingested, stored, and prepared for consumption for both exploratory and operational usage.
– Algorithms: Businesses can transform data into actionable intelligence and a set of commands, typically with a mix of business and engineering methods (e.g., identification of physical factors contributing to a machine’s deterioration) and advanced mathematics (e.g., anomaly and normality detection). The algorithms are cyclically tuned and retrained to trigger progressive improvements in the precision and relevance of the results and ensure continual adaptation to the evolving conditions of their environment.
– Consequences: When functionally is compatible, the results of the algorithms are passed directly to existing transactional applications. In specific cases, dedicated business process functionalities could be required (e.g., more granular and dynamic maintenance planning that could not be supported by traditional applications).
IIoT will create many opportunities that mainly affect three important areas (B2B applications, business processes, and business models) and take into account key challenges to be addressed (e.g., analytics, interoperability). Companies should pay special attention to the fact that effective IoT solutions require defining business models to direct real-time needs in a predictive way and ensuring interoperability in terms of technological convergence and integration of different data sources.
Establishing a strong IIoT platform able to support multiple business use cases and categories of assets requires the integration of multiple vendors (e.g., industrial companies, IoT-specialized companies, telco players, etc.) and components at the various layers of the stack, potentially even considering geographical implications (e.g., for wireless communication).
At this point in time, establishing a strong IIoT platform able to support multiple business use cases and categories of assets requires the integration of multiple players and components at the various layers of the stack. Major adopters have the unique opportunity of both reaping massive business benefits well in advance of the competition and driving the shape of the offering to make it more consistent with their long-term needs.
IoT encompasses a multitude of distinct use cases that all bring different benefits to the organization. IT and business leaders need to identify the set of IoT use cases most relevant to their verticals and then accelerate through the use journey to focus on the highest business impact areas (or IoT 2.0), which typically involve extensive use of analytics and integration into application platforms.
The key use case that stands out in this area is predictive maintenance. In many ways, predictive maintenance represents the Holy Grail of what IoT can achieve in transforming businesses and industries. This is because it has both internal and external benefits.
Things (e.g., equipment, sources, etc.) in every industry break down, often at the worst possible times or in the most inconvenient places. If you can anticipate these breakdowns, however, you can prevent them from occurring at inopportune times and in awkward places. In effect, you can avoid having the production line go down in the middle of, for instance, Christmas holiday order production because a sensor was covered with dust or a motor needed rewinding.
Predictive maintenance depends on IoT-enabled resources to capture and communicate information about these resources. That information can then be analyzed in near-real-time using predictive analytics to anticipate when the next failure will occur or, preferably, when your company needs to fix a problem before the next likely failure.
Railway infrastructure is a good example of diagnostic-platform and predictive-maintenance innovation because it’s a critical asset for developed and emerging economies, helping to ensure and improve the mobility of people and goods.
In order to ensure safety and efficiency, infrastructure operators have invested in various kinds of diagnostic technologies aimed at improving the ability to create intelligence around the current and prospective status of the assets. Such technologies include ground diagnostics and specialized trains equipped with measuring devices. Data generated by these diagnostic technologies are then analyzed with dedicated tools. While these dedicated diagnostic solutions help significantly to improve the performance of network management, the evolution of technologies in the areas of IoT and Big Data is creating the opportunity to significantly redesign the overall approach around diagnostics to unlock a new phase of maintenance innovation for railway infrastructures based on:
– Volume: Big Data is important for its statistical value. No single transaction matters. High volume is required to distinguish significant signals from noise. What is a significant signal and what is noise depends on the model you use. Models change and adapt; what is now noise can become a signal with a different or better model.
– Velocity: Streams from the field require reactions in sub-second frames.
– Veracity: Data are not true or false, rather they are usable or useless. We need to think in terms of probability and confidence intervals.
– Variety: This is the output of diagnostic equipment.
Everything is important in an IoT ecosystem, but while the infrastructure will deliver a “working” IoT solution for the business, the data side – including existing platforms, operations technology systems, analytics, and so on – will lead to true business transformation.
IoT is both complex and vast, yet essential to business transformation. It is not the sole remit of the IT department but requires key sponsors and stakeholders in the business as well.
There must, therefore, be a sound business reason for developing an IoT solution in the enterprise. The success or failure of IoT solutions (and the broader topic of digital transformation) in the enterprise hinges on the stages of IoT evolution; the defined governance model; the IoT business plan; an innovative mindset, and optimized operations.
This article first appeared at digitalistmag.com
About the Author
Chiara Martini is a project and program manager and business transformation manager lead at SAP. With 15 years of experience in managing solutions projects, she helps SAP customers navigate project activities on digital transformation in industries such as financial services and transportation. Chiara applies her well-developed strategic and leadership skills to deliver innovative solutions and methodologies.0