Industry 4.0 Product & Expertise

EBIW - Your ideal partner in business transformation

Industry 4.0 Product & Expertise

Content

Industry 4.0 Vision & Reference Architecture

• Industry 4.0 AI & ML

• Industry 4.0 eBIW Product & Expertise

• Q&A – Wrap up

2

Why change now?

Manufacturing is Transforming

3

Data Rich SiloedPerfectCustomer Aware

Insight-Driven ConnectedFastCustomer Led

Transform the

business, placing the

customer at its heart

Use data to make

decisions that improve

customer experience

Accelerate the pace of

innovation, working

with the customers,

partners and the supply

chain

Break down internal

barriers, with a focus

on the customer and

not the org chart

The Digital Enterprise

4

Manufacturers Are Focusing Outside In, With Customers At The Center

From Grease to Code, They Are Learning How to Differentiate with Digital

Detect• Track movement

• Read temperature

• Gauge humidity

• Sense vibration

Shop Floor Applications

Asset Monitoring Production Monitoring

Fleet Monitoring Connected Worker

IT Applications

Revenue and Billing Manufacturing

Logistics Service

IoT Devices

Building Sensors Equipment

Logistics Mobile Devices

Device Signals

Analyze• Visualize status

• Contextualize events

• Predict failures

• Trigger alerts

• Update device parameters

Act• Dispatch service

• Reroute shipments

• Substitute materials

• Re-plan supply

• Initiate billing events

Business Events

Focus on business outcomes

Industry 4.0 Information FlowBetter products, operating efficiency, happy customers

Integration&

GovernanceStructured Data

UnstructuredData & IOT Data

Predictions & ActionsPatterns & Correlations Genealogy & Trace

ERP

CRM

HCM

SCM

MES

QualityLIMS

Manufacturing Data HouseData Management, Preparation & Contextualization

T&A

Machine Learning & Other Data Science

Manpower Machine ManagementMaterial Method

Insight Models Predictive Models Feature Significance Models

Model Deployment

Contextual Data

Model Definition

Manufacturing Data

Model Training Model Performance Evaluation

Root Cause Analysis Impact AnalysisProactive Management

Time series Images Maps TXNs Parametric & bin

Data Types

Equipment Sensors

Scada

MQTT

OPC-UA

Data Historian

Product sensors and Test

Environmental

Industry 4.0 - Connected Machine, Method and ProcessConvergence of Operational Technology (OT) with IT Systems

Industry 4.0 - Connected Machine, Method and Process

Realtime Data Streaming

Batch Streaming

AI/ML Models Realtime Anomaly Detection

Using AI/ML Models

Realtime Alert/Notification/API

Predictive Analytics for

Condition Based MonitoringInitial & Retrained

Industry 4.0 – Architecture (AWS)

Build-Train-Deploy-Retrain-ML Model & Real-time Insight Driven Action


Realtime Data

Streaming

• Continuously generates

manufacturing process event

data in real-time which

describes something that

occurred at a given time.

• The real-time data helps the

machine learning models to

identify anomalies across

different processes and

variables.

• The data can be aggregated

with other data points in a

single real-time stream

processing data pipeline.

• Easy translation and

communication between

diverse devices, applications

(new and old), and integration

technologies.


Realtime Anomaly Detection

• Monitors production processes

continuously and detects anomalies

using artificial intelligence, machine

learning, and statistical process

control methods.

• Finds top features contributing to a

process failure using artificial

intelligence and machine learning on

historical data.

• Detects anomalies on real-time data

using the statistical process control

method.

• The data-driven approach can be self-

adaptive, it has a high dimensional

learning capability and is, by nature,

deployable on many different

application cases (i.e. machine types).


Predictive Analytics for

Condition Based Monitoring

Using AI/ML

• Real-time monitoring of industrial

equipment performance using IoT,

IIoT, and Artificial Intelligence.

• Automated scheduling of

maintenance processes and

procurement requests for parts.

• Fixing the wear and tear before the

point of failure of the asset.

• Reduced chances of downtime by

50%.

• 40% increase in production/service

delivery.

Predictive Maintenance Explained

Real Time monitoring Predictive Maintenance



Realtime Useful Life(RUL)

Prediction Using AI/ML

The method used to calculate RUL

depends on the kind of data available:

• A known threshold value of a

condition indicator that detects failure.

Degradation models estimate RUL by

predicting when the condition

indicator will cross the threshold.

• Lifetime data indicating how long it

took for similar machines to reach

failure

• Run-to-failure histories of machines

similar to the one you want to

diagnose


Real-Time Alerts/

Notifications/APIs

• Lightweight, out-of-the-box

mobile alerting solution that

addresses urgent and

targeted communication

needs of operations and

maintenance teams out on the

shop floor.

• It creates a direct pathway to

modern Industry 4.0-style

communication and

complements your SCADA,

MES, ERP, and IoT platforms

in minutes with critical mobile

alerting capabilities.

• Manage timely availability of

teams(on-call duties, shift) to

route alerts automatically.

Machine Learning and AI Embedded in

Smart Manufacturing

Statistical trend and pattern

detection for Business KPIs

Prescriptive Actions based

on root cause analysis

TREND

DETECTION

RECOMMENDATIONS

ANOMALIES

PREDICTIONS

Auto Detect deviation

from normal behavior

Predict future state by

learning from machine

data and business context

Actionable

Analytics


MQTT

BrokerConsumer &

Publisher

Cloud Gateway Stream

Processing

Azure Functions

HTTP REST

End PointWeb Portal/ Mobile App for

User Input/ Configurations

Smart Welding – An eBIW’s

Product Offering (HW & SW)


Routing Layer

Data Acquisition Layer

Machine/ Transducer Layer

Product Offering & Application

Connected

Enterprise

Connected

asset

Connected

worker

Yield Predictive

Maintenance

Quality

Insights for quality

improvement

Defect rate reduction

Intelligent Analytics and Gamification Layer

Production & OEE

Management

Autonomous

MaintenanceUtility Management

Cloud Layer

IOT Gateway Data Conversion and shaping

Key Deliverables - MIS

1. IMPROVING OPERATIONAL EFFICIENCY

2. OVERAL EQUIPMENT EFFECTIVENRSS (OEE) ENHANCEMENT in Following Domain:PRODUCTIVITY: Measurement of actual production against Norms or Captivity in real time. Suitable Management action to track the CAUSE of short-fall (if any) and initiate immediate counter measure to improve machine output close to ideal OEEQUALITY: Checking Whether Output Parameters are within the Set Limit or not. This will enable shop floor to achieve target with quality conformance. Management will be assured about shop process consistency. Same Data can be used to generate SQC parameters throughDigitised QC system. This can be used as Sales & Marketing tool to demonstrate to it’s customers. COST: Cost of operation can be drilled down to every level and can be used to minimise wastage. This will help the organisation to determine Lean KPIs for operation and ensure continual improvement of Cost optimization.

3. DIGITAL PREDICTIVE MAINTENANCEMACHINE HEALTH : In place of routine periodic maintenance and replacement of components, this system will enable Predictive maintenance of machines resulting in major saving in operational cost a t the same time ensure maximum uptime and prevention of unwarranted breakdown.

4. PLANT UTILITY

Real time monitoring of plant utilities like water, power, oil and gas will help reduce plant operational cost.

5. ASSET TRACKINGa.Enables tracking Asset Location & activity on real time basis. Useful for Construction/ Mining industries.b.Can unleash maximum asset payback by renting out during non productive hours/ period.

Deliverables

• Creates a Real Time Dashboard to facilitate LEAN implementation in operation to

improve OPERATIONAL EFFICIENCY every hour by checking actual data with

BENHCHMARK DATA and improve on the same.

Plan-Do-Check-Act- CYCLE

• Ensure M/C uptime & minimise spares cost by Predictive maintenance.

• Provide Asset sanity.

• Linked with SCM will ensure JIT delivery system of Consumables & Raw Materials.

• Quality incidence data will be used to monitor the jobs where rigorous post inspection

will be needed.

• Variation in the Output data will lead to replacement of Wear Parts proactively.

AI/ML Expertise – Platform & Solution

Data Management

Database – Data Lake - Access – Integration

- Preparation

Infrastructure

CPU – GPU – Storage -

Network

Model Training

Projects

Open Source Libraries

Model Deployment

JupyterLabNotebook

Model Management

Model Explanation

Data ScienceLibrary Auto ML

ModelCatalog

Data Access

Data Transformation

Data Visualization

Model Training and Tuning

Platform & Methods

Communications &

Media

Industrial

Manufacturing

Retail & Consumer

Goods

Airlines &

Transportaion

• Predictive Analytic for

Loading pre-paid cards

• Best-Fit Offers for Postpaid

Customers

• Revenue Leakage

• Churn Propensity

• Customer Segmentation

• Customer Fraud Detection

• Heat-zones of High Traffic

and Network Insights

• Smart Network Constraint

• Real-time Anomaly

Detection &

Visualization Product

Innovation

• Predictive Analytics for

Condition-based

Monitoring (COBM).

• Real-time Useful Life

(RUL) Prediction

• Right Product Mix &

Promotional

Effectiveness

• Flexible Supply by

demand signals and

sentiment.

• Customer

Segmentation

• Customer Sentiment

Analysis

• Market-Basket

• Employee Basket

• Employee Combination

• Product Category Mix

• Product Affinity

• Shrinkage

• Up-selling & Cross

Selling

• 360° customer profiling

• Delay Prediction and

Impact

• IATA Compliant Real-

time Baggage Tracking

• Data Monetization

• Identifying Ancillary

Revenue opportunities

• Up-selling and Cross

Selling

• Heat maps to visualize

the connection between

retail sales and departure

gates

AI/ML Expertise – Platform & Solution

Delivery of Industry Specific Use cases

Full Stack Development Expertise

• Front End Frameworks

• Angular, React

• Backend Development

• Java, Springboot, Nodejs, Python

• Databases

• Oracle, Mongo DB, Elasticsearch, Postgresql, Mysql, Dynamo DB

• Mobile Development

• React Native, Ionic

• Dev Ops / CICD

• Docker, Kubernetes, Jenkins, Git, Cron,

• Lambda, EC2, Cloudfront, Route53, Cognito, S3

A Must for Making Industry 4.0

Into a Reality

Retail & Manufacturing Comms & Media FSI AirlinesConstruction

Consultoría

Entretenimie

nto

Customer References

Industry 4.0 Product & Expertise

Documents