Getting started with IoT for manufacturing
The Industrial Internet of Things is quickly being adopted in the manufacturing industry and beyond. Gartner predicts that the Internet of Things will comprise 20.4 billion connected “things” by 2020. McKinsey predicts that the potential economic impact of IoT in factories could be between 1.2 and 3.7 trillion dollars by 2025.
The benefits of IIoT are clear: manufacturers see uptime increases of as high as 50%, cost reductions up to 30%, and user engagement improvements up to 50%.
However, the rate of failure for new IIoT projects is notoriously high. According to a 2017 Cisco report, about 60% of IoT initiatives don’t make it past the proof of concept (PoC) stage. Only 26% of companies reported that they considered their IoT initiatives to be successful. This is not surprising, as IIoT initiatives are complex transformations that impact a company’s business models, technology, architecture, processes, and people.
Here are 8 common pitfalls for industrial IoT projects and how to avoid them.
1. The initiative is not driven by a business outcome
A successful industrial IoT initiative should be driven by a business goal rather than an IT outcome. Solving IT problems or hopping aboard the “digital transformation” bandwagon won’t be sufficient drivers for an IIoT initiative. Without specific, measurable business outcomes, digitization efforts are bound to fall short.
IIoT projects should address specific business problems, which can include (but are not limited to) improving quality, increasing machine utilization, and driving faster improvement cycles. The most important element is that the goals should be measurable by KPIs and pre-defined measures of success. Without a foundation in key business metrics, the new technologies a company implements will fail to live up to its promise.
2. The business goals and/or timeline are not realistic.
It’s equally important to make sure your business goals (and timeline) are realistic. It’s wise to be conservative in estimating costs and timeline and under-promising and over-delivering in terms of savings. Use a KPI or other pre-defined measure of success when assessing the feasibility of your goals.
Make sure all stakeholders, including workers and managers, have a consistent, well-articulated understanding of the initiative’s objectives. Misaligned expectations will profoundly impact the project’s outcome. Establishing objective metrics for success is critical to ensuring stakeholder alignment.
3. Management is not aligned with the goals of the initiative
An IIoT project should be seen as a company-wide digital transformation. As such, alignment throughout the organization is crucial. Any successful business transformation requires a cultural change in order to support and sustain the initiative. Support and commitment from upper-level management are critical to guide the project. This is especially important through any rough patches the project may endure. Management should uphold consistent priorities, as internal resistance is bound to arise with any change.
4. Internal teams are not aligned
IIoT projects involve several teams managing complex, interdependent components such as hardware, device software, protocol stack implementation, backend systems, end-user applications, and more. It’s essential for these teams to function smoothly in order for the project to succeed.
Clear and consistent communication is key to sustaining a successful cross-functional team. To ensure cohesion, establish clear roles and responsibilities within the project. Consider also implementing a formal process for resolving issues.
5. The IT and OT teams are not equipped to deal with the complexity of the implementation
Implementing an IIoT initiative requires a significant redesign of a company’s IT and OT architecture. The new digital architecture will comprise many connectivity technologies, devices and applications, and management platforms. This introduces an increase in complexity that requires significant expertise. Prepare the IT and OT teams for these new, complex challenges.
This complexity also gives rise to the problem of interoperability. According to McKinsey, interoperability is required for 40-60% of the total potential economic value of an IoT implementation.
Process improvement from an industrial IoT implementation depends on being able to read data from different sources. However, each of these protocols speaks a different, proprietary language. The protocols must be able to be understood across these languages in order to bridge data silos.
6. The initiative does not adequately take security vulnerabilities into account
In 2019, cybersecurity is top of mind for those in the IIoT sphere. According to Cisco’s 2018 cybersecurity report, manufacturers have been particularly vulnerable to threats. This is due to their proprietary networks and lack of security within plants. However, newer generations of IoT hardware and IT networks, as well as cybersecurity systems tailored for IoT networks, are becoming widely available. When you choose an IIoT solution, make sure its security features are adequate.
7. There is not an adequate process for designing and testing
It’s critical for any new system to be subjected to rigorous testing prior to implementation to determine whether specifications are met and whether the system is secure. Manufacturers should ensure ample room and process for process redesign and testing.
8. There is not a plan for scaling the initiative
Manufacturers should be strategic about their goals for their projects in the PoC stages and beyond. A proof of concept, also known as a proof of value (PoV), is an experiment that should answer the following questions:
- What value does technology create for your company?
- What is your return on investment?
When designing a proof of concept, it’s important to identify distinct processes where marginal improvements can translate into much larger returns.
A proof of concept doesn’t have to be a fully scalable setup. However, it’s important that the PoC validates the project’s concept. Be sure to incorporate feedback from stakeholders about the PoC into plans for the full implementation.
As manufacturers move from the PoC to the full implementation, it is helpful to have a high-level roadmap. This provides clarity for the project, links actions to vision, and provides a reference for timeline and cost.
It’s not uncommon for industrial IoT initiatives to prove much more difficult than expected. The good news is that solutions are becoming readily available to address these specific pain points. Tulip’s platform is specifically designed to make IIoT plug-and-play and easy to integrate with the rest of your factory’s ecosystem. The no-code platform makes IIoT initiatives accessible for manufacturers, even if they don’t have IT expertise.
For a low-risk way to experiment with industrial IoT, try Tulip’s Factory Kit.