The curious case of IoT: Unlock new value from data with AI
- Industry 4.0 News
- Read 92 times
Data is the new currency. And actionable data is vital to future competency. These are among the conclusions reached by experts who participated in round table discussions hosted by ISA Watson IoT.
A series of ISA Watson IoT round tables took place in Delhi, Mumbai and Hyderabad in May. They were led by IBM subject matter experts (SMEs) Sanjeev Agarwal, Harsh Kumar, Rohit Kumar and Vinod Boggarapu, and attended by operations and CIOs from many large organizations.
Those who were present at the events acknowledged that high quality, actionable data is critical, and they posed many questions about the adoption of IoT. I’d like to share with you the issues that were discussed at this event, and that will also be addressed at our upcoming IoT Exchange in Sydney.
1. Data is the cornerstone of any IoT strategy. How do we ensure we have the right data in the right place? And how do we rationalize security concerns?
- Quantity and quality: While organizations have set themselves on a path to capture data, they want to ensure they are capturing all the relevant data to derive meaningful analytics. Experts advise that a business standpoint rather than a technology lens should drive the approach. After you have the business goal in mind, you need to break that down into sub-components to understand what data is relevant and the best way to capture the same. It can take between six and twelve months to have enough relevant data to predict outages and guide data driven decisions. As for the quantity of data, both prescriptive and predictive will require significant amounts of data. Getting the connected layer in place with real-time alerts will also improve the situation, subject to the specific use case.
- Security: Security remains a top priority. Resiliency and business continuity need to be built into the transformation agenda. With the convergence of IT and operational technology (OT), the firewall that separates the two becomes more complex and less well defined with IP-enabled devices.
Often, the CIO, whose primary concern may be network security, owns the IT organization. The person in charge of the OT might be a field operator, like a foreman, who is more concerned with performance. As IoT emerges, it will be increasingly important for both sides to focus on the same security goals without making the networks and systems difficult that it interferes with the functionality.
2.What is the gestation period before we have significant data? Also, how do we handle compatibility issues when data arise from different sources and formats?
- Data to initiate: Start with your ERP or historic data which by default will be a couple of years old. The prescriptive models on an average require at least 1+ years of data.
- Compatibility: IBM systems are flexible to accept a wide array of data sources. A Design Thinking workshop is effective to examine sources of disparate data and standardization and agree upon customization.
IT professionals participated in stimulating discussions.
3. What is the ROI that justifies the investment for IoT?
The genesis of seeking the right technology is usually a question about efficiency, outdated technology, risk aversion or expense reduction. For example, business owners often ask one of these questions:
- How can I work smarter, cheaper and faster?
- Can I automate inefficient manual processes?
- Does aging technology or environmental wear and tear put my industrial assets at added risk?
The ROI for IoT manifests itself as a combination of many factors.
- Save time, operational expenses and mitigate worker-risk: Automation supports a predictive maintenance model that prevents unnecessary downtime and replenishment of spares for machinery that is not in need of servicing. Further, this may also infuse efficiencies like improved output, reduced time to market, and even reduced workplace injuries. As an example, imagine placing sensors in a hazard prone area such as an oil well or a mine to sense conditions to send data instead of manual monitoring.
- Insights can guide business decisions: Analysis of data gathered from IoT can help improve productivity. Optimal workloads, energy consumption, best configurations and orientation for moving machinery for optimized output are some use cases that find relevance in an industrial setting.
- Enhance customer acquisition through experiences: One of the best ways to attract customers is to create a popular commodity, or become known as a cutting-edge company. IoT can be a tremendous differentiator— from smart carts that provide a rapid check-out and an improved shopping experience to sensors for water bottles that send a signal when it’s time to hydrate. If a delivery company can improve delivery times and service their customers more efficiently, the customer retention and positive word-of-mouth feedback alone could be worth the IoT investment. As for the magnitude of impact, a lot depends on the baseline. For example, 1 percent of USD 10 billion is huge, but a 10 percent improvement over current levels should be a reasonable target.
4. What are the consequences associated with shifting cost and coexistence with incumbent systems?
IBM systems can coexist with a wide variety of vendors so a total upheaval is not necessary. Taking baby steps from design workshops to piloting on subsystems can build confidence to modernize systems and leverage the full benefits of the technology.
A series of round tables were held in May 2019.
5. What is the impact on human personnel?
When does the system gain complete domain knowledge? The goal of AI at the current stage of maturity is to augment human intelligence. It is to aid a human being’s unique creativity and abstract thinking with high processing speeds, analytics and domain knowledge gained through analyzed data, best practices, known issues and resolutions, and constantly changing regulations. It’s an era of augmented intelligence for most areas and will continue for at least a decade. Some of the most mundane and repetitive tasks have and will see replacements.
6. What skill-levels and culture are needed to address new technology?
Skills remain a challenge, so organizations need to take a realistic view of what they have in-house and where they need to outsource or use partners. With the advent of AI-based computing comes a need for different set of skills. Traditional IT teams and field workers need to be re-skilled, trained and counseled for next-gen technologies. Culture is pivotal on the innovation road map. The culture of the organization and change management can define the success of the deployment.
7. Does IBM use the data to make its own systems more intelligent?
As IBM Chairman, Ginni Rometty has pointed out repeatedly, we believe the customer owns their data and the analytics derived from it. IBM will never use their data to enrich its systems without explicit approval from the customer. Clients in fields like healthcare do realize that it’s for the common good to share some level of data. This would be applicable for areas like cancer research. That apart, clients in manufacturing and allied fields may be willing to share best practices that enhance the industry as whole. Core competencies and intellectual properties are the property of the clients, for the clients’ intended purpose only.
8. Off-the-shelf Original Equipment Manufacturers (OEM) and automation are already available. Why IBM?
OEM systems with silos put interoperability at risk. Also, dealing with multiple-vendors is an operational nightmare. To address this, IBM stays at the forefront of technology innovation, earning more patents than any other organization. Further, IBM has demonstrated its commitment to advances in IoT technology by investments of USD 3 billion. With domain expertise, access to technology SMEs, and a thriving business partner ecosystem, IBM is the obvious choice. Further, IBM is proud to be a Leader in The Forrester WaveTM: Industrial IoT Software Platforms, Q3 2018. Check this report.
Clearly, IT has entered the next phase of transformation; it is not just an enabler, but the business itself. This blog addresses some of the concerns faced by businesses. However, please feel free to connect with me to discuss any concerns you might have.
To continue the conversation, please join us at IoT Exchange in Sydney, 23-24 July 2019
I would like to extend an invitation to business leaders facing similar challenges to IoT Exchange in Sydney on 23-24 July 2019. It’s a great opportunity to engage in discussions with IBM staff, business partners and customers, and network with your peers. You’ll participate in two full days of learning about new technologies through 40 information packed sessions. The curriculum has been divided into three distinct academies that will present how AI, IoT and agile technology can unlock the full value of your data: TRIRIGA® Academy for facilities operations, Maximo® Academy for asset management, and Engineering Academy for software and systems engineering. I hope to see you there!
To nominate yourself for a FREE PASS register here.