9 digitalicon-manufacturing Manufacturing

Manufacturing

[Manufacturing6] HSSE incidents

Submitted by admin on Thu, 08/18/2022 - 15:22

HSSE incidents

[Manufacturing5] Lack of quality control on manufacturing lines

Submitted by admin on Thu, 08/18/2022 - 15:13

Lack of quality control on manufacturing lines

[Manufacturing4] Higher operational costs due to lower levels of automation

Submitted by admin on Thu, 08/18/2022 - 13:18

Higher operational costs due to lower levels of automation

[Manufacturing3] Siloed data leads to ongoing operational inefficiencies

Submitted by admin on Thu, 08/18/2022 - 13:09

Siloed data leads to ongoing operational inefficiencies

[Manufacturing2] Inability to move towards autonomous plants

Submitted by admin on Thu, 08/18/2022 - 12:25

Inability to move towards autonomous plants

[Manufacturing1] Unplanned downtime impacting production

Submitted by admin on Thu, 08/18/2022 - 11:39

Unplanned downtime impacting production

5G for Manufacturing

Supercharging Industry 4.0

5G for Energy
5G for Energy

Pain points in the Manufacturing industry

Thumbnail

Pain Point 1

Unplanned downtime impacting production

Explore Solutions
Thumbnail

Pain Point 2

Inability to move towards autonomous plants

Explore Solutions
Thumbnail

Pain Point 3

Siloed data leads to ongoing operational inefficiencies

Explore Solutions
Thumbnail

Pain Point 4

Higher operational costs due to lower levels of automation

Explore Solutions
Thumbnail

Pain Point 5

Lack of quality control on manufacturing lines

Explore Solutions
Thumbnail

Pain Point 6

HSSE incidents

Explore Solutions
icon

How 5G-enabled digital solutions can help

Improve asset availability through forecasting early fault detection in critical components and conducting proactive replacements.

Data flow

Step 1: Devices

Step 1: Devices More Information
  • Sensors deployed on all critical operational assets
  • Purpose: Measure pressure, speed, flow, etc.

Step 2: Connectivity

Step 2: Connectivity More Information
  • Time series data transport
  • Asset data (maintenance records) access

Step 3: Edge Compute

Step 3: Edge Compute More Information
  • Not time critical at present (predictions timeframe are in days), may change when real-time prediction is required from other equipment

Step 4: Cloud Compute & Storage

Step 4: Cloud Compute & Storage More Information
  • All data from the edge sensors (both historical and real-time)
  • Enterprise storage is company-owned

Step 5: Applications & Services

Step 5: Applications & Services More Information
  • Predict future asset failure using ML models, initial improving model
  • PaaS-based set up, not company-owned

Step 6: Inform Decision Makers

Step 6: Inform Decision Makers More Information
  • Only reporting potential future failure so action can be taken in time

Step 7: Support Decision Making

Step 7: Support Decision Making More Information
  • End of process

Application logic

Scenario 1

  • Identify critical assets such as engines, fridges, belts, compressors, etc.
  • Collect event and/or timeseries data from these critical assets, potentially using sensors.
  • Collect as much data as possible from selected assets and surroundings as ML models will be strengthened with more data.
  • If there is a lack of data, look at options of creating and using synthetic data.

Scenario 2

  • Timeseries data will be stored long-term in Enterprise storage.
  • AI (ML) application requests all the data it needs for ML model development, from Enterprise storage to copy in its own storage cache.
  • Development of the ML model is done through an iterative process and a quality ML model (fully data-driven) will require multiple steps to detect anomalies/potential failures.
  • SME involvement working with data scientists is required to develop the model.
  • Models will be stored and maintained by AI applications.

Scenario 3

  • The process from data collection to execution of models is fully automated.
  • Only alerts to operations in case of anomalous data behavior (e.g., resulting in expected asset failure in X days).
  • Operations takes action to replace suspected assets in time, therefore avoiding unplanned downtime.
  • ML model management is important to handle large number of models.
  • Various options for visualisation of entire operations (e.g., XR options on OpenXR platform).
icon

Expected benefits

  • Proactive detection of abnormalities and preventative actions
  • Improve productivity from reduced unplanned downtime
  • Reduce maintenance cost due to early spotting of issues
  • Shorter turnaround times due to more targeted maintenance

Note: 

iconAI – Artificial Intelligence

iconML – Machine Learning

iconMV – Machine Vision​

iconSME – Subject Matter Expert

iconP&ID - Piping and Instruments Diagram

iconGHG - Greenhouse Gas

iconIT - Information Technology

iconOPAF - Open Process Automation Forum

iconOT - Operational Technology

icon

Key value created

  • Increased availability of the plant translates to increased productivity

Email us at:

enterprise@digital-nasional.com.my

Phone number:

0123456789

Potential

COGNITE

global.rokid.com

COGNITE

test 2

COGNITE

test 3

icon

How 5G-enabled digital solutions can help

Decisions will be increasingly made directly on the basis of the AI outcome, starting with low-risk decisions and evolving over time to high-risk decisions.

Data flow

Step 1: Devices

Step 1: Devices More Information
  • Broad spectrum of sensors collects a wide variance of data (images/flow/temp/ speed,etc)
  • Sensors: fixed/drones/ robotics (driving/swimming)

Step 2: Connectivity

Step 2: Connectivity More Information
  • Imagery, time series, event, etc. data transport

Step 3: Edge Compute

Step 3: Edge Compute More Information
  • Storage and compute, in some cases, at the edge level for time-critical decision making
  • Can be for any type of data

Step 4: Cloud Compute & Storage

Step 4: Cloud Compute & Storage More Information
  • Non-critical data stored in Enterprise (cloud) storage
  • Critical data to be copied from Edge to Enterprise storage

Step 5: Applications & Services

Step 5: Applications & Services More Information
  • MV and ML applied to assist in decision making from simple to — over time — advanced tasks
  • Autonomous means that AI makes the call

Step 6: Inform Decision Makers

Step 6: Inform Decision Makers More Information
  • Final decision made by AI process (becomes increasingly complex over time)

Step 7: Support Decision Making

Step 7: Support Decision Making More Information
  • End of process

Application logic

Scenario 1

  • Data (event/timeseries/images/ etc.) collected from various sensors (e.g., temperature, pressure, etc.) – currently present or to be installed on machinery.
  • Assume a broad spectrum and large numbers of sensors.
  • Cameras collect high resolution video and image data to be accounted for in analysis.

Scenario 2

  • 5G/edge compute will enable AI to process time-critical data to take proper actions in real-time which will be needed for various actions.
  • ML and MV processed data beyond prediction, building rule-based business logic for automated decision making.
  • Iterative process for developing these models, to achieve required accuracy (mandated step wise approach).

Scenario 3

  • Incremental approach – automated decision making will be applied for low-risk tasks and will increase in risk level until complete autonomous operation is achieved (will require time).
  • Real-time explainable and trusted decisions using complex business logic that is traceable to understand the logic flow of AI.
icon

Expected benefits

  • Timely and data-driven decision making
  • AI-generated forecasts and scheduling, optimising costs (e.g., repairs)

Note: 

iconAI – Artificial Intelligence

iconML – Machine Learning

iconMV – Machine Vision​

iconSME – Subject Matter Expert

iconP&ID - Piping and Instruments Diagram

iconGHG - Greenhouse Gas

iconIT - Information Technology

iconOPAF - Open Process Automation Forum

iconOT - Operational Technology

icon

Key value created

  • Accurate automated decision-making increases operational efficiencies

Email us at:

enterprise@digital-nasional.com.my

Phone number:

0123456789

Potential

COGNITE

global.rokid.com

COGNITE

test 2

COGNITE

test 3

icon

How 5G-enabled digital solutions can help

Relevant, real-time data sources are aligned in enterprise data storage to enable accurate and timely AI based decision-making.

Data flow

Step 1: Devices

Step 1: Devices More Information
  • Cameras collect images, video, meter readings, etc.
  • Sensors and other devices will collect flow rates, pressure, temperature, etc.

Step 2: Connectivity

Step 2: Connectivity More Information
  • All data types collected will be transmitted
  • Different protocols may be utilised

Step 3: Edge Compute

Step 3: Edge Compute More Information
  • Not required, may change for certain equipment where failure lead times are shorter

Step 4: Cloud Compute & Storage

Step 4: Cloud Compute & Storage More Information
  • All data stored in Enterprise storage
  • Keys not aligned between multiple storage: Time series; Asset Maintenance; Engineering; P&ID, etc.

Step 5: Applications & Services

Step 5: Applications & Services More Information
  • Applying Digital Twin-aligned technologies to align these critical data sources such that asset information can be located immediately

Step 6: Inform Decision Makers

Step 6: Inform Decision Makers More Information
  • Should greatly improve MOC (Management of Change) activities and related turnaround work

Step 7: Support Decision Making

Step 7: Support Decision Making More Information
  • Step towards Digital Twin

Application logic

Scenario 1

  • Critical data sources are identified across the operation of the plant.
  • Point cloud/3D engineering models as input.
  • Data could be received from many different sources (real-time and non-real time).
  • Real-time sources will increase over time.

Scenario 2

  • Locate the keys (unique identifiers) in use across data sources.
  • Use Digital Twins-based technologies to match the various data sources to create a single 3D-aligned view with all data collected in view.
  • For matching, various technologies will be used.
  • Using this single view to do engineering/maintenance-related activities.
  • Quality of matching improves over time.

Scenario 3

  • Drives improvement of quality of the data sources and therefore matching as well.
  • Will have a large positive impact on any maintenance job in terms of speed and quality.
  • Apply AI on completed projects to determine how further improvements can be made.
  • SME involvement working with data scientist is required.
icon

Expected benefits

  • Speed of execution for any maintenance/engineering project greatly improved
  • Speed & quality of MOC (Management of Change) and turnarounds improved

Note: 

iconAI – Artificial Intelligence

iconML – Machine Learning

iconMV – Machine Vision​

iconSME – Subject Matter Expert

iconP&ID - Piping and Instruments Diagram

iconGHG - Greenhouse Gas

iconIT - Information Technology

iconOPAF - Open Process Automation Forum

iconOT - Operational Technology

icon

Key value created

  • Various aligned datasets will enable better and faster decisions, resulting in increased uptime

Email us at:

enterprise@digital-nasional.com.my

Phone number:

0123456789

Potential

COGNITE

global.rokid.com

COGNITE

test 2

COGNITE

test 3

icon

How 5G-enabled digital solutions can help

Maximise usage of 5G-enabled robotics in various parts within manufacturing lines, data collection for remote monitoring, maintenance, etc.

Data flow

Step 1: Devices

Step 1: Devices More Information
  • Sensors supporting critical business workflows impacted
  • Sensors: fixed/robotics/drones
  • Different types of sensors

Step 2: Connectivity

Step 2: Connectivity More Information
  • Imagery and time series data transport
  • Asset maintenance data
  • Other data types

Step 3: Edge Compute

Step 3: Edge Compute More Information
  • Real-time decision impact made at the Edge storage and compute
  • Mixture of real-time and non-real time decisions

Step 4: Cloud Compute & Storage

Step 4: Cloud Compute & Storage More Information
  • Historical video and sensor data from multiple cameras
  • Loaded and kept in Enterprise storage

Step 5: Applications & Services

Step 5: Applications & Services More Information
  • Machine Vision (MV) and Machine Learning (ML)
  • Different ways to inform the operator

Step 6: Inform Decision Makers

Step 6: Inform Decision Makers More Information
  • Alarm-based reporting when damage, wrong meter readings, etc. are detected

Step 7: Support Decision Making

Step 7: Support Decision Making More Information
  • End of process

Application logic

Scenario 1

  • Decide in the following sequence:
  • Business workflows to be impacted
  • Data input needs
  • Sensor (camera/robotics/drones/etc.) to be installed; frequency of data flow
  • Exact location and connectivity of sensors
  • Is the response time critical? If so, then use Edge-based and if not, then Cloud-based 🡪 This trend is expected to move towards real-time decisions.

Scenario 2

  • All data loaded into Enterprise store and then offered to the AI solutions as needed.
  • AI solutions to be ML/MV-based:
  • ML: Spotting changes in behavior
  • MV: Spotting changes in appearance: Damages; Corrosion; Meter readings; Switch settings; etc.
  • Above will run on Edge (when time critical) or on cloud.
  • All data will be stored in Enterprise storage.

Scenario 3

  • Developing these ML/MV models will be interactive and will not be ready off the shelf 🡪 SME + data scientist is required for the development where SMEs are often from the enterprise.
  • All will operate automatically end-to-end.
  • Make a clear list of business workflows and implement the list step by step.
  • Cameras should have 720P image quality.
icon

Expected benefits

  • Reduce reliance on human workforce and physical surveys through remote monitoring
  • Early detection of faults to be rectified
  • Increase system availability

Note: 

iconAI – Artificial Intelligence

iconML – Machine Learning

iconMV – Machine Vision​

iconSME – Subject Matter Expert

iconP&ID - Piping and Instruments Diagram

iconGHG - Greenhouse Gas

iconIT - Information Technology

iconOPAF - Open Process Automation Forum

iconOT - Operational Technology

icon

Key value created

  • Timely identification of faults reduces downtime impact

Email us at:

enterprise@digital-nasional.com.my

Phone number:

0123456789

Potential

COGNITE

global.rokid.com

COGNITE

test 2

COGNITE

test 3

icon

How 5G-enabled digital solutions can help

Automated quality control employing machine learning with sensors collecting data and alerting during abnormal production for improvement purposes.

Data flow

Step 1: Devices

Step 1: Devices More Information
  • Sensors to be added to monitor quality-related data
  • Type of sensors driven by type of man: cameras (various); specs to monitor; etc.

Step 2: Connectivity

Step 2: Connectivity More Information
  • Equipment sensor data transport
  • Energy availability data (e.g., steam, solar, wind, etc.)

Step 3: Edge Compute

Step 3: Edge Compute More Information
  • It is expected that the majority will be real time-based, so direct intervention is possible
  • Defective products to be separated in real-time

Step 4: Cloud Compute & Storage

Step 4: Cloud Compute & Storage More Information
  • Equipment data stored to train AI algorithms
  • Storage is Enterprise-owned

Step 5: Applications & Services

Step 5: Applications & Services More Information
  • AI (ML/MV) applied to detect goods anomalies in various forms and provide insights on quality control

Step 6: Inform Decision Makers

Step 6: Inform Decision Makers More Information
  • Alert-based notification of quality anomalies

Step 7: Support Decision Making

Step 7: Support Decision Making More Information
  • End of process

Application logic

Scenario 1

  • Install relevant sensors with focus on real-time intervention since that has the best financial impact.
  • Make a list of the quality aspects in order of importance then decide what sensors are needed.
  • With 5G, the amount of bandwidth needed is not a constraint. Therefore, take this into account when selecting sensors.
  • Cameras should have 720P image quality.

Scenario 2

  • SME and data scientists are required and will work as a team.
  • It is expected that Edge will be the prime home for AI in this case.
  • Ensuring that all data always goes to the Enterprise store.

Scenario 3

  • Develop the AI (ML/MV) models highly interactively to get quickly to the right quality level for this job.
  • Link to industry quality standards when developing these models.
icon

Expected benefits

  • Automation increases quality control sample sizes and improves accuracy of detection
  • Detection allows early rectification of production workflow and therefore reduces wastage

Note: 

iconAI – Artificial Intelligence

iconML – Machine Learning

iconMV – Machine Vision​

iconSME – Subject Matter Expert

iconP&ID - Piping and Instruments Diagram

iconGHG - Greenhouse Gas

iconIT - Information Technology

iconOPAF - Open Process Automation Forum

iconOT - Operational Technology

icon

Key value created

  • Detection of quality issues in real-time for increased productivity

Email us at:

enterprise@digital-nasional.com.my

Phone number:

0123456789

Potential

COGNITE

global.rokid.com

COGNITE

test 2

COGNITE

test 3

icon

How 5G-enabled digital solutions can help

With 5G-enabled sensors monitoring behavior around the plant to track, manage and predict potential occupational hazards in real-time.

Data flow

Step 1: Devices

Step 1: Devices More Information
  • Cameras (fixed/drones/robotics/people) and sensors capture real-time behavioural, asset and facility insights

Step 2: Connectivity

Step 2: Connectivity More Information
  • Video, imagery and sensor data transport

Step 3: Edge Compute

Step 3: Edge Compute More Information
  • Time-critical: Edge AI camera will identify anomalous behaviors and defects in facility.
  • Other data is stored in the cloud

Step 4: Cloud Compute & Storage

Step 4: Cloud Compute & Storage More Information
  • All collected data stored in Enterprise storage

Step 5: Applications & Services

Step 5: Applications & Services More Information
  • AI applied to detect potential threats and hazards and ensure compliance (PaaS)

Step 6: Inform Decision Makers

Step 6: Inform Decision Makers More Information
  • Alert-based notification of suspicious activities

Step 7: Support Decision Making

Step 7: Support Decision Making More Information
  • End of process 

Application logic

Scenario 1

  • Cameras, sensors and edge devices installed throughout the facility to gather high-resolution video and imagery data.
  • Drones are deployed to increase coverage (including hard-to-reach areas) and transmit live data through 5G.
  • Camera coverage drives the quality of the HSSE coverage.
  • Cameras should have 720P image quality.

Scenario 2

  • AI (MV) models created using the data gathered to manage infrastructure, staff and operational risks.
  • Data processed by AI at 5G edge generates real-time virtual fencing for automated occupancy management (e.g., lifting of goods).
  • Machine vision (5G-based) can detect real-time potential threats.

Scenario 3

  • Automated compliancy to safety standards such as OSHA, IOGP 577, CCOHS.
  • Actionable insights generated can provide decision makers with recommendations to prioritise specific tasks.
  • AI-generated inspection checklists and emergency response plans.
icon

Expected benefits

  • Worker safety (real-time monitoring & hazard alert)
  • On-site worker safety compliance/PPE/cost reduction
  • Increased productivity (reduced hazards/accidents)

Note: 

iconAI – Artificial Intelligence

iconML – Machine Learning

iconMV – Machine Vision​

iconSME – Subject Matter Expert

iconP&ID - Piping and Instruments Diagram

iconGHG - Greenhouse Gas

iconIT - Information Technology

iconOPAF - Open Process Automation Forum

iconOT - Operational Technology

icon

Key value created

  • Reduced number of safety incidents and related work absenteeism with improved staff morale

Email us at:

enterprise@digital-nasional.com.my

Phone number:

0123456789

Potential

COGNITE

global.rokid.com

COGNITE

test 2

COGNITE

test 3

Cross-industry solutions

Autonomous robots and aerial drones can traverse harsh and dangerous environments and reduce hazard risks for human workers. Such technology will provide enhanced monitoring and safety upkeep for facilities. They can be fitted with various sensors geared towards the purpose they are meant for and by using 5G, we can deploy them in real-time whereby the data collected can be sent directly to an AI application for investigation and immediate further action.

Wearable technology allows security and medical personnel to conduct background and health checks efficiently. Of course, they have a much broader focus and can be used in many different industries where clothing has been enriched with wearable technologies. Wearables are also able to collect extra information for a 5G-enabled digital service.

Sensors are crucial for capturing real-time data and insights and pave the way for more informed decision-making. Sensors will be the data collection point for every 5G-enabled digital service and therefore will have to cover a broad spectrum of types of information to be collected and the way they are connected to the DNB 5G network.

XR is set to take work and play to the next level with cutting edge technology never seen before. These devices have the potential to boost productivity and enhance entertainment, changing the way we experience our reality through digital transformation. There is an increasing broad spectrum of devices available for both consumer and enterprise markets, making it increasingly important for each user experience.

DNB Ecosystem

Disclaimer of Endorsement

All information provided on potential 5G-enabled solutions and potential vendors is provided for information purposes only and based on our awareness of the available 5G-enabled solutions and potential vendors in the market. It does not constitute endorsement, recommendation or favouring by DNB. It is your responsibility to verify and evaluate such 5G-enabled solutions and potential vendors. DNB is not an agent or legal representative of any of such 5G-enabled solutions and potential vendors, is not associated or endorsed by such potential vendors, has no authority to act on behalf of any such potential vendors, and will be an independent party if you enter a business relationship with such potential vendors. Third party links included are for convenience only and are not under DNB’s control. DNB does not assume any responsibility or liability for your use of such third-party links.

Subscribe to Manufacturing