A component in the machine listening for new updates and incorporating those updates to the DB placed in the machine

Kibana dashboard is created as a starting point. This issue will work as a thread to capture the feedback on dashboard having discussions about what to add/delete.

Following are the list of attributes which need to be sent to the cloud in (near)real time:

START TRANSACTION FORMAT

"TIMESTAMP", "Machine_ID", "FLAG", "Batch_ID", "Line", "Batch_Number", "Recipe", "Set_Point", "Tolerance", "Speed", "e", "T1SP", "Prd_Cost", "Lbr_Cost", "Extr_Cost", "Operator_Qt", "Scale_Qt", "Batch_Qt", "Line_Leader"

END TRANSACTION FORMAT

"TIMESTAMP", "Machine_ID", "FLAG", "Batch_ID", "Line", "Recipe", "Total_Packs", "Aggregated_Wght", "Total_GA", "Avg_Speed", "KPI", "Avg_Wght_End", "AvT1", "Run_Time", "Pause_Time", "Prd_Cost_End", "Ga_Cost", "Lbr_Cost", "Extr_Cost_End", "Total_Cost", "Pack_Cost", "Line_Leader"

SCALE TRANSACTION FORMAT

"TIMESTAMP", "Machine_ID", "FLAG", "Batch_ID", "Scale", "Op_Name", "Set_Point", "Avg_Wght", "PPM", "T1PPM"

Access control must be applied on the data being stored at the cloud. Access Control Diagram shows the hierarchy to be implemented.

• No data access across clients
• A client must be able to access all sites data
• Distributors have access to all clients' Engineering data relatable to them
• PAL has access to all the data
• Where to incorporate division of Production and Engineering data? (Question for Rui)

As per the discussions with Rui on 22/04/2021, following changes need to be made in the config file:

• Average Weight in END transaction should have a unique name that can help to implement SUM function at the Kibana dashboard. Currently, Average Weights of SCALE and END are being added.
• Serial Number attribute should be changed to Machine ID
• Flag attribute to be added (Three different flags for different Production datasets)

Following attribute will be sent to the cloud on event basis. Latency tolerance for this data is yet to be decided.

"TIMESTAMP", "Machine_ID", "FLAG", "Alarm Description"

The landing page of the Mi Weigh dashboard will be having various visualizations of the data for the last one month (to be discussed). This timeframe will be editable by the user and upon selecting a specific timeframe, all the visualizations would be updated accordingly. There should be a way to further filter data w.r.t Lines.

• By default, landing page contains the data of last one month without any further filter applied
• Input field to select the timeframe to fetch the data from
• Data filter ability w.r.t Production Lines

Following are the attributes which will be sent to the cloud in (near)real time:

"TIMESTAMP", "Machine_ID", "FLAG", "User", "Product", "Recipe", "Tool", "Cycles", "Total_Packs_Prod", "Run_Time_Prod", "Idle_Time_Prod", "Alarm_Time", "Next_Machine", "Printer"

To be filled later

Following changes will be made in Kibana dashboard as per the feedback:

• Redirect from the whole row batch data table, currently it requires selecting column 3 (Landing Page)
• Limit decimal points of Cost per Pack in table (Landing Page)
• Remove the visualization titles (Both)
• Average Weigh metric to be fixed
• Min-Max values of average weight graph to be shown
• Remove 1st and 3rd visualization graphs
• Update sample production data to incorporate Flags

Following are the data attributes for audit data which is being generated whenever changes are made in machines. This is also an event based data and latency tolerance is yet to be decided:

"TIMESTAMP", "Machine_ID", "FLAG", "User Group", "Description", "Value Before", "Value After"

As referred in #5, the landing page batch lists must enable the user to drilldown to a detailed view. This detailed view of batches must have the following visualizations related to the selected batch data:

• Production speed graph showing details of Packs per minute produced over the complete batch
• A graph showing average weight over the complete production batch
• A graph showing production speed w.r.t the line leaders
• Total number of packs produced
• Cost per minute of the batch

For Mi Weigh Dataset, the landing page must have a table populated by list of batches satisfying the date and line filters. There should be a way for the user to zoom into the detailed view of batch. Description of the detailed batch view are provided in a separate card.

• A table in landing page populated by list of batches where each row contains Batch ID, Timestamp, Cost per packet, Recipe and Line Leader
• User is able to go to the detailed batch view by selecting any row