"We cannot solve our problems with the same thinking we used when we created them."

Albert Einstein

Synthese Real Time Performance Management

SYNTHESE REAL TIME PERFORMANCE MANAGEMENT®

Synthese Real Time Performance Management® is a modern and unique continuous improvement model that can be implemented as the first layer of any managerial structure in a production management environment, in any type of industrial organization that wants to efficiently exploit the many possibilities of big data and the concepts of digitalization. 

In our model, we transform in real time all your available data into meaningful and contextualized information, to then update and prioritize it dynamicaly and finaly distribute it immediately to the critical nodes of all your processes.

Our model allows a highly optimized use of the strengths and resources of your entire organization, to achieve a sustainable control of your performance. 

We reduce to nothing the typical resistance to change, which is replaced by a continuous accumulation of positive results of small changes.

Our complete management model is fully implemented after the completion of five phases :

– Phase 1 : DATA TRANSFORMATION

– Phase 2 : DYNAMIC VISUALIZATION OF DEVIATION

– Phase 3 : NODE COMMUNICATION

– Phase 4 : DYNAMIC SENSITIVITY ANALYSIS AND PRIORITIZATION

– Phase 5 : IN DEPTH METHOD COACHING

All pictures and graphs below are example of a managerial system based on the  Synthese Real Time Performance Management® model and implemented in a hot rolling mill.  All visual information has been made available to everyone on company intranet and to key operational people on specific screens in the workshop.

Example of a Standard Production Process

In order to illustrate our concept, let’s figure out an online production process, several machines in different workshops gradually add value to your products.

Equipment (machines) are generally controlled by PLC (programmable logic controllers) or electrical or electro-hydraulic controls. Those local control systems are generally named as level 1 (LEVEL 1).

The entire production process is controlled by an MES (Manufacturing Execution System), which we qualify as level 2 (LEVEL 2).   This system links the different machines together and communicates with all of them to ensure tracking and synchronization of operations.

On top of that, we may also find a system that manages communication with the outside world (sales, marketing, purchasing). It will be named as level 3 (LEVEL 3).

During normal operations, those systems interact with each other, exchanging information. Level 1 systems may receive little information to operate, however can generate multiple information in the heart of their automation systems.  This information exchanged is always technical data, technical setting or adjustment, component or material tracking, energy consumption, temperature, pressure, torque, forces, triggers, etc.

If those information allow to steer and monitor the progress of production operations continuously, it is very often at the end of a campaign, or a day, or a shift that we can examine the “performance” of the whole process.

Our model “Synthese Real Time Performance Management®” allow to know and visualize process performance in real time and to act at every moment on that process performance. Now let’s see how those many real time data are collected and used in our “Synthese Real Time Performance Management®” model. Let’s have a look to the same production process online.

We collect in a dynamic database the maximum of this information exchanged in real time between the different equipment (LEVEL 1) and the monitoring systems (LEVEL 2 and LEVEL 3). When information is absolutely necessary for the evaluation of a performance, but is not exchanged and remains at the level of the PLCs of a machine (LEVEL 1), we connect either the database directly or the MES (LEVEL 2) via a local interfacing card. It may be necessary to transform analog data into digitized information via conversion cards.

The frequency of collection depends on the type of production. If one product goes through the whole line every 20 seconds, we collect all the data attached to that product, so every 20 seconds.The collection frequency is therefore 180 batch of data per hour.

A first work is carried out on the incoming data in order to eliminate disturbances, linked to measurement errors, shutdowns, connection problems, etc. This is the REAL TIME DATA PROCESSING.

These data collected and “cleaned” in real time feed a database, where they are stored entirely for an one to three months period, this is the SHORT TERM DATA STORAGE database. This database will be used to have off-line all the information necessary to completely reconstruct the performance graphs of any production campaign that happened in the near past, in order to be able to carry out in-depth phenomenon analysis.

Those data will then be stored in a more compact format in a permanent database, the LONG TERM PERFORMANCE & REFERENCE database. We will find there all stratified and averaged information allowing to calculate references which are useful for the model with the PERFORMANCE REFERENCE PROCESSING module.

The real-time data and the reference data of all performance parameters are then used in a GRAPHIC VISUALIZATION PROCESSING module in order to produce the visualization graphics in real time. The refreshing frequency of those graph must be smaller that the data collection frequency to show all changes in real time.

A final calculation engine, the SENSITIVITY PROCESSING, allows the calculation of sensitivities in real time and supplies the same graphics engine, for updating and displaying the VALUE DRIVER TREE.  It is at this level of calculation that we dynamically (therefore variable over time) highlight which deviations from the respective references cause the most significant impacts on the overall performance, and that is indicated in the VALUE DRIVER TREE.

Therefore, we know at any time on which part of the process or sub-process requires an immediate correction or action. 

PHASE 1 - DATA Transformation

Transform your real-time data in actionable insights

All existing real-time data available at all stages of the process and all levels of the organization are transformed in pre-designed actionable information available to all, mainly into the form of visual graphical performance information using simple visualization codes and continuously updated real time.

Those transformed information speak the shopfloor language and allow workers and team to better understand their influence on the process parameters.

PHASE 2 -DYNAMIC VISUALIZATION OF DEVIATION

Continuously compare current performance to references

All data relating to performance are continuously stored into a dedicated database. Multiple references are automaticaly calculated, stored and layered over time depending on the existing process variability.

Typical value are annual averages, however, time frame can be adapted to the capacity of resources (Quarterly, Monthly, moving average…).

Each and every process parameter is then displayed into a easily redeable compact preformated graphical information, where gaps between current performances and their respective references can be immediately highlighted .

Example 1 : evolution of production rythm (Beam Blank per hour) over a whole production campaign. Graphical information is clearly understandable, impact of numerous stoppages on final productivity after a very good start of campaign.

Example 2 : evolution of production rythm (Beam Blank per hour) over a whole production campaignwith mix of two Beam Blank sizes with different reference (annual average productivity), the mixed reference is continuously recalculated.

Example 3 : evolution of production rythm (Beam Blank per hour) over a whole production campaign. Despite roll breakage after 3 hours, final figures exceed references.

Example 4 : evolution of production rythm (Beam Blank per hour) over a whole production campaign. Despite very hectic start of campaign, final figures exceed references.

PHASE 3 - NODE COMMUNICATION

Communicate contextual adapted information to critical nodes of the process

Beside standard technical and operational information which are shown and made available though HMI (Human Machine Interface), our model adds contextual information relating to the performance of the current process.

This additional performance related information is critical and helpful for operators and team leaders to make the link between their actions or decisions and their influence on performances of all processes they manage.

Example 1 : All real time graphs showing the complete performance status of the rolling mill are available in the Supervisor Cabinet.

Example 2 : A set of four contextual graphs showing required information for the Saw Team Leader is available in the Hot Saw Cabinet.

Example 3 : A set of four contextual graphs showing required information to properly manage the performance of the Reheating Furnace is available in the Reheating Furnace Team Leader Cabinet.

Example 4 : A set of four contextual graphs showing :

– Torque  at Rolling Stand 1

– Beam Blank temperatures (model & measure)

– Furnace Productivity in BB/hr

– Hot Saw Efficiency in Cuts/hr

this specific grouping of information helps RHF Teamleader to accelerate rythm with a perfect understanding of risks of roll breakage and bottelnecks at Hot Saws.

Example 5 : A set of four contextual graphs showing :

– Torque  at Rolling Stand 1 (most critical pass)

– Torque  at Rolling Stand 2 (2 passes)

– Torque  at Rolling Stand 3 (3 passes)

– Torque  at Rolling Stand 4 (3 passes)

this information enables the rolling mill supervisor to control the effects of the diameters and wear of the rolls and of temperatures on the torques, in order to redistribute the rolling forces over the various passes.

Example 6 : Contextual graph showing specific Gaz consumption at Reheating Furnace (nM3/ton), combined with Beam Blank productivity (BB/hr).

this information helps the RHF supervisor to act on the regulation of the zones of the reheating furnace according to chosen discharge rates and the rolling stoppages in order to optimize gas consumption.

PHASE 4 - DYNAMIC SENSITIVITY ANALYSIS AND PRIORITIZATION

Dynamicaly value impact of deviations to reference on the global performance

Finally, in our DYNAMIC VALUE DRIVER TREE (VDT), we continuously quantify real time the impact on the global performance of all positive or negative deviation measured for every process or sub-process parameter you decided to follow. 

The most significant gaps to reference are highlighted dynamicaly.  They SYSTEMATICALLY show where immediate action is required to improve performance, or at least get back to standard.

As it continuously allows to fix the most critical deviation, our VDT is the most powerfull tool to fully optimize the usage of your resources.

PHASE 5 - in depth model Coaching

Leverage the strength of your entire team for a long-lasting performance

Those rapid decisions taken by the operators closest to the job – and having the best understanding of the decision at hand – lead to tangible and sustainable improvements, while enhancing at the same time deep commitment across the team.

Gradually, the role of the leaders and managers will evolve into an more organizational one: create an situation of empowerment where the workers can act and fail safely, guided much more by their own influence on performance and less by procedures.

This builds up professional capabilities and lift business understanding at all levels of the organization while this gives the workers a strong sense of ownership but also fosters a preference for action.

Synthese helps and guides you to implement our advanced and unique proprietary continuous improvement model.  It allows your teams and processes to reach their full potential by optimizing resources and making full advantage of all available data.

+352 28 99 22 02

+352 661 530 667

SYNTHESE S.A.
9 rue basse L-4963 Clémency
info@synthese-management.com
RCS Lux B246912
TVA LU32410354