DISCRETE EVENT SIMULATIONBASED ON REALTIME SHOP FLOOR CONTROL

DISCRETE EVENT SIMULATION-BASED ON REAL-TIME SHOP FLOOR CONTROL 21 st European Conference on Modelling and Simulation (ECMS) June 4 th - 6 th, 2007, Prague / Czech Republic Presenter: Franck FONTANILI Authors: Samieh MIRDAMADI, Franck FONTANILI and Lionel DUPONT Department of Industrial Engineering/Ecole des Mines d’Albi-Carmaux/France

Summary 1. Shop Floor Control [SFC] 2. Discrete Event Simulation [DES]-based on SFC 3. Manufacturing Execution System [MES] for SFC 4. Requirements evolvement into on-line simulation 5. Experimentation Plate-Form and Out-Line

1. Shop Floor Control [SFC] a. b. c. d. SFC definition & its sub-functions A classification of SFC Control tools to aid decision-making Problematic of SFC Orders Information Constraints Objectives • Shop order priority Production Control F. O. Launching Information Flow F. O. Reporting SFC • Planned decision Court or middle Term Tactic All the activities of short-term production in agreement with the objectives established by the production control, by adapting the production to the disturbances which can occur on the level of the workshop. [APICS, 05], [Grabot, 97] (Order / Monitoring) Consign Actions Row material Information feedback Operation System • Real time data info. • Resources efficiency • WIP quantity information Real Time Operational … • Shop order status info. Goods …

1. Shop Floor Control [SFC] a. b. c. d. SFC definition & Its sub-functions A classification of SFC Control tools to aid decision-making Problematic of SFC Exploitation Off-Line Control On-Line Control Experience feedback Predictive Control To Preparation Estimated data Existing System Before Execution Very weak probability of events occurrence System in the course of execution Unforeseen events and critical drift of system variables (e. g. cycle time) Proactive Control Reactive Control To Anticipation To Correction Experience feedback Temporal axis

1. Shop Floor Control [SFC] a. b. c. d. SFC definition & Its sub-functions A classification of SFC Control tools to aid decision-making Problematic of SFC Product life cycle Conception Re-engineering Exploitation CAD, CAM, CAE (DESIGN, MANUFACTURING, ENGINEERING) CAPM, ERP Scheduling CAPE (PRODUCTION ENGINEERING) Specific to the Production Systems MIS, MES, CIM, CAIT Supervision Order / Monitoring (API) MES Manufacturing Execution System (often coupled with the supervision) DES Tools for Discrete Events Simulation of flows To Design To Improve To Control

1. Shop Floor Control [SFC] a. b. c. d. SFC definition & Its sub-functions A classification of SFC Control tools to aid decision-making Problematic of SFC DES MES Allows to anticipate in the future Brings a lot of information allowing to make decisions but does not allow to direct connection to a real system in the course of execution but does not allow to make sure that they are the good decisions. Use On-Line Discrete Events Simulation for the decision-making aid in Reactive and Proactive Control of workshop by coupling with MES

2. DES-based on SFC Objectives Predictive Control Proactive Control Reactive Control capacity Machine of the a. SED Application in production breakdow queues… b. Different use of simulation • n Determi (unforese nist or • E. g. en but strong Cycle known) probabilit time … y of Drifts of • Very appearanc a weak e (cycle machine probabilit time…) (unforese • y of Off-Line en and occurrenc Simulatio unknown e)… n • Off To Applicati Line minimize on (non the drifts Simulatio direct compared n connectio to the Applicati n) deadlines on (non envisaged direct

a. SED Application in production b. Different use of simulation 2. DES-based on SFC Predictive Control Process Off-Line Simulation Measure Decision Validation Reactive Control Process Monitoring Filtering Evaluation On-Line Simulation Optimization Correction Proactive Control Process Identification Off-Line Simulation Optimization Validation Stocking

3. MES for real time Shop Floor Control APS ERP Differed Time Management I. S. q Executes the commands (orders) of the production control. Supervision Order / Monitoring Fabrication process q Delivers relevant Real Time Fabrication I. S. Operational Decision MES Tactic Strategic Information information on the follow-up and the realization of the shop orders in real time.

4. Requirements evolvement into On-Line Simulation I. Validation of the simulation model II. On-Line connection III. Data usable: IV. Availability and Correctness of the data V. Data Acquisition VI. Classification and events analyzes VII. Initialization of the model VIII. Response time (Speed of the simulator) IX. Correction of the parameters in real time

4. Requirements evolvement into On-Line Simulation I. Validation of the simulation model § Connect the model to the expected or received reality § Relevance of the model: sufficient quality § Completed of the model: all the necessary information § Establish the fidelity of the simulation model § Produce a definitive proof to support the model § Reduce the risk II. On-Line connection Simulation Model OPC MES Data Base Ethernet networks of the real system API … API Supervision

4. Requirements evolvement into On-Line Simulation III. Data usable: Availability and Correctness of the data Data usable Availability complete Correctness Incomplete Measure Errors Way to collect IV. Data Acquisition § Use of a data base of workshop § Methods of Acquisition: Sensors: measure by Detector DBMS: Information system Up to dateness Application

4. Requirements evolvement into On-Line Simulation V. Classification and events analyzes Events Unforeseen Foreseen Unknown Known Strong Probability with lapse of time Weak Probability On-Line appearances Unknown probability VI. Initialization of the model • The real system must be planned from the current state of the system. • To start with, a “not-empty” state of the model corresponding at the real state.

4. Requirements evolvement into On-Line Simulation VII. Response time (Speed of the simulator) • Simulation time Reasonable response time between decision-making and control execution t 1+ t t Real Time • t depends on the characteristics of the workshop VIII. Correction of the parameters in real time • To transfer and execute the best realizable solution from simulation towards MES

5. Experimentation Plate-Form and Out-Line Operational System Information System Ethernet networks of the real system D. B. M. E. S Networks Sensor Actuators OPC Server Loading API Histories data OPC Client Unloading API Operation API Transfer API Real Time Data Simulation Model (SIMBA)

5. Experimentation Plate-Form and Out-Line § Experimentation of reactive control on the Technology plat-form § Collect data of the ground by MES § Filtering of the events releases § Initialization of the simulation model by injection of the collected data § Possibilities of application to a logistic chain § Performance analyze

Thank you for your attention

a. Off-Line Simulation process b. On-Line Simulation process 1. DES-based on SFC Shop orders To realize Predictive Control Before Execution Off-Line Simulation Results analysis Achieved objective? yes No Validation How will it take place during the execution in case of appearance of an unforeseen event…? Reactive Control During Execution Optimization of the Control parameters Launch of the execution of the shop orders on the real system ? Realized Shop orders

a. Off-Line Simulation process b. On-Line Simulation process 1. DES-based on SFC Predictive Control Realize the estimated parameters by predictive control Reactive Control During Execution Application of the proactive control result Shop orders execution In progress Known events? Scenarios of The current state of the system Real-time Simulation Correction of the control parameters on the model Yes Initialization of the simulation model Activate the On-Line Simulation Validation (Without modification) Yes Achieved objective? No Yes No Filtering the events to release simulator Events release ? Case Base Classification of the unforeseen events Monitoring of the real system state No Proactive Control yes Variation on objective? No Realizable decision-making in real time Correction of the control parameters on the real system

Predictive Control Before Execution Shop orders To realize Estimate the most frequent disturbances Off-Line Simulation Diverse scenarios of Off-Line Simulation Results analysis yes No Optimization of the Control parameters Reactive Control During Execution Prepare the relevant solutions Pooling Validation Achieved objective? No Yes Launch of the execution of the shop orders on the real system Realized Shop orders Case Base Results of Proactive Simulations Proactive Control Before execution a. Off-Line Simulation process b. On-Line Simulation process 1. DES-based on SFC