Artificial Intelligence is benefiting to various industries including healthcare, education and manufacturing. But what is Artificial intelligence (AI)? In Layman language, a simulator of human intelligence, which makes the decision after analyzing various data utilizing a collection of different intelligent technologies including machine and deep learning, analytics and computer vision.
The fourth industrial revolution is employing AI to enhance its overall efficiency. The technology is not only helping to reduce manufacturing cost as well as it is improving productivity and quality.
Artificial Intelligence in Manufacturing: How Manufacturers are Reducing Cost and Improving Productivity?
Manufacturing is a capital-intensive process, and once a plant is a set-up, replacing, removing or renovating is exorbitantly expensive. New machines improve performance; reduce redundancies, while improving overall quality metrics. AI is proving an alternative route to achieve all this and at extremely competitive price points.
Instead of now replacing machines, manufacturers are adding AI/ML tools to pre-inspect raw materials identify defects, perform quality evaluations, and a lot more. Many AI solutions currently deployed use vision, which provides an opportunity to further add features like correct labelling, batch numbering, size/dimension mismatches, as well as other visual inspection elements.
Rapid Advancements in Robotics in both shopfloor & warehouse are driving new levels of Operational productivity. Manufacturers can expect improvements in overall effectiveness up to 25 %.
Nilesh Auti, Global head of Manufacturing, Tech Mahindra
Cost Reduction in Manufacturing
The very fact that one can monitor results at each phase of the manufacturing process and enable one to take corrective measures at the initial stage, saves a lot of investment that was usually spent on rectifying damaged products or machinery problems. Besides, AI also ensures predictive maintenance of machines and optimize usage of assets within a specific production unit.
5 Ways AI is Reducing Manufacturing Cost:
5 Ways AI is Reducing Manufacturing Cost
# Predictive Maintenance is using Advanced AI aid in drastic reductions in downtime costs by 15%.
# Extending the Remaining Useful Life (RUL) of production machines & equipment through predictive Maintenance.
# Computer Vision based AI algorithms can help assess real-time Worker Productivity &detect imperfect operations & train workers
# Throughput improvement of 3% or higher through Intelligent Scheduling, Prediction& control of process Parameters, Bottleneck prediction etc.
# Automatic pipette systems and robotic arms, liquid handling robots
Enclosed are few case studies for your reference
# Paint shop Optimization through AI Based Auto sequencing helped improve:
= Throughput by 3%
= User confidence by 70%(adherence)
# Remote Monitoring for Leading Motor Manufacturer:
- Improving equipment effectiveness
- Minimize business disruption
Brake Life prediction – Predict Remaining brake life through Analysis of Brake failure patterns & helped airlines with:
= 12% cost savings through RUL recommendation
= 20% reduction in unscheduled Maintenance
How can Artificial Intelligence Increase Productivity?
AI enabling manufacturers to make quick and error-free decisions. It can help to make decisions for any fault that can occur in future, in advance, so you can take appropriate action even before it can damage your production.
The operational effectiveness of a manufacturing unit is closely dependant on three main factors –
AI technologies help to manage the availability of machines, material and manpower (3Ms) through improved prediction and monitoring. Efficiencies of the operations can be improved through real-time observation of operational parameters using sensors and immediate decisions based on insights from these parameters using AI technologies.
AI can predict the quality of output by analysing several hundreds of parameters – both core and ambient – and narrowing down the key factors that impact the final quality of the product.
AI also has a key role to play in improving other aspects of manufacturing – such as power consumption or effect effluent treatment. Such improvements bring down the overall unit cost of manufacturing without compromising the overall quality and timely delivery of the product.
How AI is Impacting Manufacturing Industry?
AI technologies will have a widespread impact on shaping the future of manufacturing. The degree of impact is closely dependant on the complexities and the operational needs of the specific manufacturing unit. The impact value increases immensely as the AI adoption evolves from Diagnostic applications towards Predictive, Prescriptive and eventually Cognitive applications.
Even the Indian manufacturing units that are usually labour intensive and cost-conscious can benefit significantly with AI technologies. Products manufactured by Tier 1 companies need to compete globally in terms of price as well as consistent quality. Adoption of AI at relevant levels becomes imperative for them to be successful. Consequently, this also pushes the need for relevant AI down the manufacturing supply chain.
Manufacturing Industry is one of the prime beneficiaries of Digital, capturing market growth of nearly $8 trillion delivering new customer experiences, process efficiencies and improving asset utilization. The future of manufacturing lies in embedded intelligence that will enable businesses to:
Innovate in operations- Operation Model Transformation: AI enables self-healing automation in factories, supply chains, and other processes through:
# AI enabled robotic process optimization/ digitization of manual processes
# Higher Levels of Automated Decision Making
Generate New Revenue Streams- New Business Models:
# Autonomous Driving & Advanced Driver-Assistance System (ADAS) features
# Personalized offer based on UniqueID
# Voice control in Vehicle
# Next generation navigation
# Maintenance scheduling
# Advanced fleet management for shared vehicles
Data Enabled Services to deliver new customer experiences:
# Enhanced flight experience
# Enhanced service –first time right through AI enabled service knowledge management
# Virtual assistants to answer customer queries
# Navigating assistants to support search and buy products on websites
Prashant Gupta, Head of Solutions, Verizon Enterprise Solutions India
With the emergence of AI, the manufacturing sector is creating safer operating environments, increasing efficiencies by 24x7 operations, predicting machine failures and repeatable tasks are being carried out by AI driven Robots.
One example is the use of automated guided vehicles (AGV) transporting pallets, cartons and products throughout a manufacturing facility. If a vehicle encounters any obstruction on the manufacturing floor, the concerned vehicle will react and move quickly.
The second example could be illustrated on how industrial robotics is used to test hazardous materials. A robotic arm connected allows users wearing a tactile glove in a remote location to feel different textures and temperatures of items on the factory floor. The goal here is improved safety. Workers can handle a spill or accident without putting people in danger.
Will AI Create or Destroy Jobs?
Next 4-5 years is going to be a honeymoon period for everyone, where AI solutions add performance and quality improvements. We are already seeing industries adding AI to their process automation. Once we cross this 4-5 year barrier we will see that AI become synonymous with Robotics and that is when jobs would be under threat. For owners, that’s cost saving, but it will have a huge overall societal impact. Robotics with AI will add unbeatable speed and efficiency, and reduced oversight.
R. Venkateswaran, Senior Vice President, IoT Solutions, Persistent Systems
It is important to clear a misconception about loss of jobs due to AI. Several studies by global organizations such as World Economic Forum have concluded that while certain kinds of routine jobs impacting about 30% of the work-force will get displaced by AI, it is equally important to note that several additional jobs will get created.
E-commerce warehouses are great examples of what we are about to see. Amazon seems to be running their whole warehouse facility on full automation, where product selection, packaging, in-warehouse transport to delivery to trucks/shipment vessels, humans are not to be seen.
Because we do not have a mature ecosystem of industrial automation, robotics and AI, there is still some time before we see problems at job levels. Most factories in India are human resource intensive. Similar factories outside India (say for an extreme example Korea), has 15 humans to one robot. In India, we only have 1 robot per 333 humans.
Rohan Shravan - Founder & Director of Inkers
Even if we were to hit Korea’s number, say, in 10 years, we will see massive un-employment, and those number are only going to get smaller. People in foreign countries have already seen high-end manufacturing, and unfortunately in India, it would be done by machines. Domain expertise and skills would be limited to only a select fortunate few.
AI has penetrated businesses across verticals- from online platforms such as food delivery and retail to manufacturers with heavy-asset operations. The implementation of AI has improved efficiency, enhanced convenience and boosted an enterprise’s ability to predict future needs at a fraction of the costs otherwise allocated to traditional alternatives. said that, “Data is the new oil and AI is the best tool to extract, process and implement this data to deliver micro-targeted, optimized business decision-making.”
One of the longest problems ailing the manufacturing sector has been the lack of a quality mechanism for proper inventory management. Due to constant fluctuations in consumer demand, there is a tendency for manufacturers to either overestimate or underestimate the amount of stock they need.
As a result, their entire product cycle suffers from constant delays, leading to client dissatisfaction or idle stock, both of which result in losses. AI has helped manufacturers walk the inventory tightrope correctly by analysing exactly how much inventory they need to keep at a particular time.
This is done by studying the market data and obtaining accurate estimates for future demand of the company’s products.
Create thousands of designs in a fraction of time – Companies such as Airbus are using AI to create thousands of designs in a fraction of the time that it would take for human teams. Technologies such as 3D printing are also helping manufacturers significantly reduce the time it takes to transport and test new products and ideas. The cost to establish multiple manufacturing centres has reduced, while logistical infrastructure, both in terms of time-efficiency and security, continuously grow due to AI.
AI Enabled machines communication – Additionally, equipment in the manufacturing industry is increasingly being controlled through distributed systems known as Machine-to-Machine (M2M). M2M technology uses sensors to record the data that is fed through these networks. Through the Industrial Internet of Things (IIoT), this data then provides important inputs to the variety of machines that are being used to perform a number of different tasks. Several manufacturing spaces today are making use of AI sensors in their gas turbines to adjust fuel valves to facilitate lower emission levels.
AI plays a vital role in M2M communication. It allows machines to communicate amongst themselves, with the operating managers and with parts of the supply chain as well. It also enables the equipment to display alerts when there is a problem or if a machine is about to break down. This facilitates proactive responses from human teams, in order to avoid interruptions that may slow down or affect operations. Hence, these practices ensure costs are reduced while there is an increase in productivity or imminent risk.
Impact of AI on Jobs: 1.8 Million to be Eliminated, 2.3 Million to be Created - Arup Roy, Research Vice President, Gartner
What will be the Future of AI in Manufacturing?
AI has just started in India to establish its foothold in this sector with small and medium-sized enterprise (SME) manufacturers still trying to decode the implication and overall economics of adopting this new-age technology. Technology upgradation is slow owing to the paucity of funds as well as lack of awareness of the importance of this technology.
The need of the hour is government provisions for intensive trainings to industrial workers on using the smart machine and AI powered tools on the shop floor. The government recently allocated Rs 6,000 crore towards the development of Tool Rooms across the country to support small manufacturing players by providing support in terms of design and quality.
Govt of India's DHI (Dept of Heavy Industry) and CII have also recently formed Samarth Udyog council, to promote digital technologies in manufacturing. One of the key initiatives is AI-ML-analytics in factory to help improve productivity. IISc, IITs and many large groups such as Mahindra, Tata as well as technology experts like Siemens are part of this body to promote AI-ML.
AI is looked upon as the propellant of fourth industrial revolution that will make knowledge-based manufacturing a reality. This includes:
Lights out manufacturing – Autonomously run factories with zero human presence & no lighting (machines can function in the dark)
Unmanned factories – Transporting equipments to unmanned transport trucks to warehouses by centralized robotic controls
Accelerated product development – Generative Design incorporating robotics in physical trial and error manufacturing process
Working at nanoscale requires precision beyond human ability and robotics will be the way forward
Digital twin enabled predictions to help in manufacturing related planning
Enhanced transport modalities by access to real time information of every taxi, bus, airplane leading to improved passenger experience and process efficiency
Digital personal/research assistants will emerge that will scan documents and compile all information
The manufacturing industry involves a lot of machinery-driven operations, which include intricate software and core-programming. However, these machines were completely dependent on humans to detect and solve even the most basic errors and technical difficulties. Through AI implementation, this drawback can be eliminated.
It is only after AI that true automation can be achieved, and even heavy machinery involved in manufacturing becomes more intuitive. Additionally, AI has allowed human teams to focus more on tasks based on value addition. Manufacturing industries can implement AI to enhance existing processes such as to improve and customise products, smart supply chains and to create new business models.
Integrating AI into operational processes in the manufacturing industry facilitates the preservation and standardisation of knowledge. This will help companies increase the productivity of their workforces, as AI has the ability to adapt and learn on its own based on large data sets and machine learning algorithms.
Thus, it is also able to deliver smart operational decisions and consistent outputs based on its accurate predictions. The benefits of this have been highlighted through the early adoption of AI by companies such as Google, Amazon and Uber, which has boosted their ability to adapt to a number of dynamic conditions and stay ahead of the curve.
In order create a more productive environment and make proactive decisions, several manufacturers are making use of AI and its related technologies such as machine learning (ML), deep learning and the internet of things (IoT). There is, thus, a greater need for manufacturing companies with volatile margins and capital-market pressures to adapt quickly in order to not be left behind the competition. For this, such companies need to be able to gauge the changing needs of their various stakeholders by utilising their data to realign processes and improve overall performance.
In near future, these robots would be able to run completely unsupervised and reach human level manoeuvrability. AI is exceedingly contributing to product compositions, simulations, process automation sequences, structural analysis, quality assessment and visual inspection. Currently, these things are still supervised, needing human intervention, and with time algorithms are getting better, and very soon be completely un-supervised.
The Massachusetts Institute of Technology (MIT) Review released its tenth breakthrough technologies of 2018 and AI was featured prominently.
One of the most intriguing AI/ML (machine learning)-led technology currently being developed is a neural network architecture called a Generative Adversarial Network (GAN). GANs operate by pitting two neural networks against each other: one network tries to simulate data samples after examining a set of real data, and the other network tries to discriminate the simulated data samples from real data samples.
This competition between networks results in a generating network that can learn the underlying structure of data in an unsupervised manner to create highly realistic simulations. Most GAN applications are currently focused on imagery, producing samples of photorealistic computers.
Analysis: AI models that can automatically learn the structure of data in an “unsupervised” manner (as opposed to today’s “supervised” approach that requires humans to painstakingly label millions of data samples) is the next frontier of AI/ML and one of the capabilities that will produce another explosion of performance gain. The technology world is actively exploring GAN use in a variety of manufacturing efficiencies.
How AI would Empower Quality Management?
Quality is imperative in manufacturing. Quality degradation may occur due to poor packaging of the product, transportation issues, machinery defects, supplier process defects in addition to metallurgical defects. Each of these defects can incur an immense loss in the form of higher customer rejection rate, high cost of production, and eventually loss of revenue for the manufacturing firm. Quality improvement has a bigger role to play in these.
In today’s Industry 4.0 world, increasing reliance on Big Data analytics in predictive manufacturing. An advanced machine learning algorithm can analyse processed data collected from the production system to provide –
a) Early warning for Process Perturbation
b) Predict Product Quality
Vikas Gupta, Managing Director, Wiley India
Using AI, manufacturers can collect data at each phase and act faster on an arising issue – be it raw material quality check, transportation or equipment performance of various machineries used.
These eventually help manufacturing firms to reduce the rejection rate of their products while keeping a constant eye on the root causes of each issue surrounding the manufacturing process.
AI enabled algorithms can notify manufacturing teams of emerging production faults (likely to cause quality issues) including deviation from recipes, subtle abnormalities in machine behavior. Aftermarket data about performance, claim history, field quality data can be used to improve quality back at production enabling faster re-engineering and visual quality inspection.
As quality assessment is a laborious process involving attention to a minute large number of details. AI through visual perception has already beaten humans in recognitions, measurements and inspection tasks. More and more companies are adding AI based tools for quality measurement, and since you can scale computers efficiently, you can increase quality inspection speed as you like. Along with AI algorithms, and robotics, camera quality has also increased tremendously.
Today industrial cameras can work at 100+ degree Celsius while capturing 250 frames per second. This means inspection can be done right at the production level, without reducing the production speed.
Because of the nature AI and vision algorithms work, every product can be tested for conformance, while on the other hand, quality inspection generally was done on few samples, extrapolating the results to the whole batch. This is a massive increase in data points, and this would only help get very accurate at quality metrics.
How is Machine Learning Applied in manufacturing?
Machine Learning algorithms are the backbone of Artificial Intelligence. In Machine Learning, algorithms are trained over objectives. These objectives could be to identify objects, read text, measure areas or volumes, determine defects, calculate joint alignments, move robotic arms based on a specific trajectory while avoiding objects, finding the orientation of objects, finding the right structural design for a component, etc. A collection of these and many more objectives would create a flow of activities in manufacturing.
For every such objective, engineers would collect the data (hence supervised) and then train a model (for example a deep neural network) to achieve the same objective. As training data improves in its quality and variation, the algorithm becomes better. These supervised algorithms area already very common these days. In a few years, we would see un-supervised algorithms replacing them, where any new variation can be automatically handled by the AI programs.
Below are factors driving explosive AI growth:
# Advanced computing power
# Increased Data Availability
# Open Source Algorithms availability
Machine Learning would help tap 20-25% growth of engineering intensive connected segment over the next five years.
Use Cases of ML includes:
$ Time series analysis and anomaly detection algorithms for identification of wear and tear and failure ahead of time and enable predictive maintenance
$ Computer vision along with other structured inputs and real time monitoring, to identify defects and deviations in the production assembly lines –
1. Reducing rejection rates of the products
2. Enhancing quality
3. Reducing the manpower requirement.
4. Pattern recognition helps subject matter experts in
$ Root cause analysis
$ Repair concessions
$ Identification of top factors affecting production parameters
$ Finding bottlenecks in production
$ Voice assisted bots workers on the shop floor to able to get speedy information and knowledge
$ Overall optimization of manufacturing processes in areas like SCM, asset management, inventory management and demand forecasting.
What are the Key Elements to Keep in Mind When Adopting AI?
Firstly, for the success of digital initiatives, it is very important for the manufacturing organizations to evolve into data-centric decision-making organization. This is a radical shift in mindset for the executives as well as the employees.
With this mindset in place, the adoption of AI should focus on relevant areas on the basis of a thorough assessment of the current state and the strategy of the manufacturing unit. The priorities of the organization are decided by the senior management and these priorities decide the overall roadmap for transformation via technology adoption.
The digital revolution requires that the key decisions related to manufacturing should be based on data generated from the core operational systems as well as other organizational systems (eg ERP, HR etc.). The value of AI adoption is strongly correlated to the quality and accuracy of the data.
Data accuracy can be ensured by collecting them as close to the source of origination – either the factory floor leveraging IoT or directly from the organizational systems through well-defined APIs (Application Programming Interfaces). Any manual intervention in this process is likely to compromise the quality of data. Historical data are a rich source for powerful analytics and can improve the quality of insights from AI systems.
Today most of the AI is confused with Machine Learning (Key Differences between AI and ML) or Supervised Algorithms. When algorithms are supervised, it is as good as the data it is trained on. One has to focus a lot on the dataset creation for the algorithms to consider all the possible variations. A good example would be - object detection. With lighting, camera distance, camera lens or conveyor belt speed changes, what camera records is changed. All object detection models would change unless the training data included all of these and yet many more elements.
Machine Learning is not a perfect solution which can handles all of this variance automatically, and one has to invest time and energy into it. But once a robust model is trained, the returns on investment is unparallel. Machine learning has in the process of improving robotics, and it is currently much researched problem. We are still 3-5 years away from running robots on AI (unsupervised algorithms). It would be a very tough decision for any manufacturer investing in robots and automation today knowing that within next 3-5 years they would be obsolete or at least 2 generations behind new ones.
4 Key elements to keep in mind when planning to adopt AI?
1. Organizations planning to adopt AI should focus on data management, security networks and AI platforms that will support their requirement
2. Clearly defined Key Performance Indicators (KPI) to assess Return On Investment(ROI) as AI is an expensive investment
3. Integration of data sources and data cleansing is critical
4. Other consideration include:
$ Dynamic and scalable storage capacity to accommodate humungous volumes of data
$ High efficiency networks at scale is necessary
$ Graphic processor units to optimize data center infrastructure & improve power efficiency
What is the Key Role of IoT in AI Deployment?
IoT is the enabler of AI would be an understatement. IoT only provides the data from connected assets but the role of making it actionable lies with AI. It offers context & creativity on IOT data thus resulting in connected intelligence. AI is far more advanced in identifying patterns & creating actions than traditional methods like post event processing or real time processing.
There are several disruptive technologies that are being introduced across industries today, all of which serve specific purposes and functions that contribute to a larger goal. IoT and AI are no exception to this, where IoT connects machines and uses data generated from them to perform functions and AI enables machines to function with intelligent, human-like behaviour.
Rajat Narang, Director, Absolutdata Analytics
IoT technology stores data that enables the remote monitoring and management of processes which are then put into action by AI in real-time. This helps improve outcomes, by providing information that helps to speed up production, and improve overall quality and processes by reducing waste.
In manufacturing, the ‘smart’ element is introduced by using IoT and AI to process data in order to help companies make informed decisions that will significantly improve production. This is carried out through sensors and machines that send this information to the cloud via IoT connected systems in the factory. The data is then analysed by AI systems in the correct context to create actionable insights to improve business operations and return on investment.
IoT and AI work together and complement each other. IoT is focused on connecting hitherto disconnected devices and entities. The data from these entities become a rich source for AI technologies to work. IoT systems ensure that the data is available in real-time and the sanctity of the data is assured by collecting them closest to the point of origination of the data.
What are the Usage and Applications of Artificial Intelligence in Manufacturing?
There is immense transformation brought in through the implementation of Artificial Intelligence in the manufacturing sector globally. One of the key concerns revolves around the ongoing maintenance of manufacturing equipments and production of line machineries incur huge expenditures due to unplanned downtime.
AI enabled robots can reveal insights on every single stage of the manufacturing procedure and later furnish accumulated data through analytics software resulting in accurate future predictions and behavioral pattern on various circumstances including manufacturing time, quantity and quality.
The application of AI as flexible and intelligent decision management solutions has the potential to transform the manufacturing sector - from engineering, procurement, supply chain management, industrial operations (production and related functions) to marketing, sales and customer services. It is among the main technological building blocks of Industry 4.0.
As per a recent article published in Wiley Innovation Black Book on Emerging Technologies 2019 – “Adopting and adapting to these technological changes is no longer an option…it is imperative for manufacturers to respond quickly to changing customer demands and maximise new market opportunities.”
Application of AI spreads across the entire gamut of the manufacturing value chain.
$ Autonomous Vehicles are synonymous with AI
$ Driver assistance solutions such as obstacle detection/avoidance, lane keep/lane departure warnings, path planning etc
$ Design optimization - 30% costs savings in design.
$ Improved asset reliability- Condition based monitoring/prescriptive analytics for maintenance of equipment machinery etc.
$ Improved workforce efficiency through CV based repair inspection, cognitive concession, optimized technician routing can be enabled through AI
$ AI enabled Smart Logistics solutions in Supply chain include demand forecasting, spend analytics, automated inventory monitoring, route optimization, remote tracking & monitoring
Example: Supply Chain Analytics – Supervised & Unsupervised ML to extract insights from supply chain data help mitigate supply chain risk
Sales & Marketing:
$ Hyper personalization, customer sentiment analytics and focused promotional offers to enable better sales conversion and increased revenues
Venkateswaran, Senior Vice President, IoT Solutions, Persistent Systems
AI can autonomously monitor and operate manufacturing units efficiently to consistently produce high quality products with a great degree of predictability.
Artificial Intelligence in Manufacturing: Ways of Deployment
AI has been deployed in a number of ways to help with manufacturing processed thereby reducing cost and enhancing productivity. Some of them are:
$ Enhanced Asset Utilization: implementing measures like Predictive (Machine Learning) maintenance has led to higher availability of critical assets & lower unplanned downtime
$ Improved Manufacturing Yield: Models leveraging machine learning are constructed using historical data to predict Yield from near real-time production data and optimized by adjusting input parameters.
Improved Manufacturing Quality: Similarly, models leveraging machine learning are constructed using historical data to predict quality from near real-time production data and optimized by adjusting input parameters so that the output satisfies customer CTQs
Enhanced Quality through Image Analytics: Models to detect flaws that otherwise would be invisible to human eye ensure better quality of product
Improved alignment of Supply & Demand through data driven forecasting. Production is flexibly rescheduled based on fluctuations in demand near real-time. The inventory is managed through better planning and real time supply chain optimization. This leads to making thinking supply chains a reality.
Improved Labor Productivity by Human-Robot (COBOTICS) collaborations. Hazardous, repetitive/ monotonous, or heavy-lifting tasks can be performed by robots enabling employees to concentrate on other tasks that are more value add in nature.
Sreenivasan V, President, Lines of Business, ITC Infotech
There is tangible benefit in greenfield scenarios using the above implementations. Additionally, brown-field situations can benefit on a case-to-case basis depending on the availability of data and the flexibility of adding or replacing a line or an asset with latest equipment and the use of robotics.
What are the Possibilities with AI in Conventional Industries such as Manufacturing or Supply Chain?
With AI penetrating the manufacturing industries of today at a rapid pace, the future of the industry is geared towards ‘Smart Factories’ and other smart spaces which make use of AI and ML algorithms to process data and streamline production processes.
Generative design is one such process that is helping designers and engineers plan and carry out the designing process by feeding it into the software. The generative design software then analyses all the data, studying every possible option to produce alternative designs. Using ML, it is able to test and learn about what is required based on the data gathered from each interaction. Thus, in this way, those responsible for product design are able to easily select and create one particular design based on all their requirements after a thorough AI machine-based analysis.
Additionally, using a function called ‘digital twins’ has also made it easier for companies to carry out remote work. A digital twin is a virtual version of a particular product, service or process that works using IoT supported by ML and AI tools. It collects information about a physical item through embedded sensors, which then send this data to a cloud-based system. Based on this, teams can make appropriate changes and improvements accordingly. Similarly, smart manufacturing also uses similar processes to help manufacturers predict and quickly address errors or problems in products and services.
Since most manufacturing businesses involve warehouses, companies are increasingly beginning to use a process called ‘reinforcement learning’ to manage their warehouses. For this, robotics-based systems are used to carry out reinforcement based algorithms, which help with managing inventory, space, and tracking transit time. This has helped companies monitor processes efficiently and ensure increased and improved production.
Prashant Gupta, Head of Solutions, Verizon Enterprise Solutions India
We are looking at a future where a factory can run end to end without any supervision. This, though would affect jobs, would help move factories out of cities and very close to raw material sources (since not many humans would be required to re-locate). Future factories would be a testament of human engineering and efficiency, where not only raw material consumption is reduced, but waste is efficiently recycled and re-used.