From Lean to Fat Manufacturing: A Tale of the Industrial Revolution

Introduction

The term "fat" is, of course, used in sarcasm here, but it is time that factories that manufacture become smarter than ever, and many already did. Welcome to my second article; I started lean, to begin with, but I hope to elucidate some of the interesting happenings in the manufacturing world – transforming a lean manufacturing system into a not-so-lean, automated one backed by Industry revolution.

To begin with, lean Manufacturing, coined by John Krafcik in 1988, primarily focuses on reducing production and response times between suppliers and end-customers. But this always comes with a compromise – issues with quality and wastage. The focus was not much on "reducing wastage" or "automating production systems" (at least in the beginning) but more on time to market.

?

Devices started speaking

But then quality and competition started to matter, and most importantly, devices started to speak – not that they did not earlier. We began to understand their language, capture their pain points, and monetize them for our use. This reinvented how businesses design, manufacture and distribute their products. Per my understanding, the focus is less on distribution and more on design and Manufacturing. While Machine Learning and Artificial Intelligence added ignition to this, more and more use cases emerged that fuelled automation and improved the quality of deliverables. This revolution demanded more use cases where data could be monetized to make the "leaner manufacturing systems" into "more productive, smart, and automated manufacturing systems".

?

Industry 4.0 – Pillars

?At AION-Tech Solutions Ltd., we focus on four major pillars of Industry 4.0

• Cloud Computing
• IoT
• Cognitive Computing
• Big Data Analytics

This article, however, focuses more on IoT Analytics, and the intended audience is mainly from the Manufacturing Industry.

Our recent joint venture with Quantron AG, a German-based e-mobility major, opened doors to venture into IoT Analytics and explore how we can monetize this effort for larger manufacturing companies across the globe.

Coupled with that are the statistics that IoT will reach ~ $950 Bn by 2026, making it an excellent opportunity for AION-Tech Solutions to innovate and contribute. According to a recent report from Accenture, the adoption rate of Manufacturing companies for Industrial automation and IoT is almost 84% across the globe. These companies believe that IoT could transform their businesses by playing a critical role in their digital transformation.

Our strategy at AION- Tech Solutions is to focus on a few key areas, such as:

• Data: Help our customers digitize their processes and monetize their key data assets for data-driven decision-making.
• Physical devices: Help our customers identify the potential to make their devices, sensors, and other physical assets start speaking.
• Use Cases: Identify potential use cases that can help them reduce outages, increase the performance of their manufacturing systems and processes
• Connectivity: Identify potential data leakages and opportunities to connect devices to make a smart nervous system.
• Real-time: Build high-performance applications that reduce time-to-act and provide opportunities to excel by capitalizing on real-time data and insights.

?

Smart Manufacturing?

Industrial automation made Industry 4.0 and Smart Manufacturing almost synonymous. IoT Application Protocols designed on open standards by the OASIS consortium brought many protocols that suit the manufacturing industry in today's demanding world.

While there are several IoT protocols, we will focus on a few by neatly categorizing them based on their applications.

IoT protocols are categorized as follows:

• Application protocols: enables network entities to identify and interact with each other.
• Industry-specific protocols: Specific to industries like Telecommunications and Electric Vehicles.
• Device-specific protocols: Protocols that help cross-vendor devices to speak and collaborate to achieve a common business problem.
• Transport protocols: Data transfer protocols define how data gets packaged and transmitted.
• Network layer protocols: Connecting one network entity to another.
• Security protocols: Protocols that support end-to-end encryption and security mechanisms.

?

A few important protocols that are commonly used in Manufacturing, especially in Machine-to-Machine data transfer, include:

• MQTT?(Message Queuing Telemetry Transport)
• AMQP?(Advanced Message Queuing Protocol)
• CoAP?(Constrained Application Protocol)
• LwM2M?(Lightweight Machine-to-Machine Protocol)
• XMPP?(Extensive Messaging and Presence Protocol)
• DDS?(Data Distribution Service)
• OCPP?(Open Charge Point Protocol)
• OBD2/CAN Bus?(On Board Diagnostics / Controller Area Network)

?

IoT and M2M Protocols

In a large-scale manufacturing setup, different devices, sensors, and equipment interact, making a complex nervous system. The problem earlier was that they used to communicate only to produce a product, but the manufacturers were not focused on utilizing the data produced by these physical components – the digitization/signals were never considered.

The logic behind using signals is simple. Questions like:

• How do you utilize signals produced by each of these devices to optimize KPIs?
• How do we capitalize on these signals to predict any equipment failure?
• How do we optimize the productivity?
• How do we ensure the safety of the workers by looking at the signals?
• How do we utilize these signals to comply with ESG and sustainability guidelines?

?

Telematics and Telemetry Systems

Our Product Development Team is heavily involved in building the Telematics System that utilizes an interlude of MQTT and Kafka systems for capitalizing data emitted by IoT devices installed in electric vehicles to create the base telematics system.

However, there are several other cases where this data could be used to predict the performance of key components of electric vehicles such as batteries, engines, throttles, motors, coolants, etc.

At AION-Tech, our Data Science practice focuses on building turnkey solutions that help Electric Vehicle manufacturers make vehicles that can offer an optimal experience to drivers and vehicle owners. Everyday use cases like State of Health prediction, State of Charge prediction, Remaining Useful Life prediction, Charging Load prediction, breakage, etc., are built in-house utilizing advanced Machine Learning and AI.

?

Manufacturing Industry

We work with some of the largest manufacturing companies (Chemical, Engineering & Construction, pharmaceutical drug manufacturers, etc.) for their end-to-end BI implementation. With our new venture into Data Science and Generative AI, we are excited to build applications that capture data, especially in the M2M environments.

Notable use cases we worked on include:

• Predictive Maintenance?– Implementing measures to guarantee seamless operation of equipment and machinery, and adeptly forestalling malfunctions before they materialize.
• Condition Monitoring?– Monitoring a parameter in the condition of machinery to identify a significant change indicative of a developing fault.
• Quality Monitoring?– Ensuring the quality of products is high and maintained by analyzing sensors and smart meters in real-time for any violations of device thresholds.
• Energy Optimization?– Identify energy waste and enable real-time adjustments to energy usage.
• Anomaly Detection?– Ensuring plant safety and efficiency by analyzing anomalies in real time.

?

AION-Tech's Advantage

Our experience in building large-scale applications and high-performing data engineering pipelines makes us a preferred one-stop Consulting / Product Development company. We worked with some of the world's leading Life Sciences, Pharmaceutical, and Electric Vehicle Manufacturing companies, making us the best fit for building such solutions.

?

?

?

?

?

?

?

?