last modified: 18-MAY-1987 | catalog | categories | new | search |

NESC9700 AVPROG.

AVPROG, Monte-Carlo Simulation of System Availability

top ]
1. NAME OR DESIGNATION OF PROGRAM:  AVPROG.
top ]
2. COMPUTERS
To submit a request, click below on the link of the version you wish to order. Only liaison officers are authorised to submit online requests. Rules for requesters are available here.
Program name Package id Status Status date
AVPROG NESC9700/01 Tested 18-MAY-1987

Machines used:

Package ID Orig. computer Test computer
NESC9700/01 IBM 3033 IBM 3084
top ]
3. DESCRIPTION OF PROGRAM OR FUNCTION

AVPROG is a large-scale simulation program used to estimate steady-state availability of a system, typically a large production unit. The system is modeled as  a success logic flow block diagram of equipment affecting the production output. AVPROG simulates system operation over a specified time period using Monte Carlo techniques. System components are failed and restarted using numerous types of distributions that are available. Among the distributions available  are the exponential and the lognormal. AVPROG computes availability  based on averages from a number of simulation cases (batches) for the specified production time interval. It also ranks unavailability contributors to permit investigation of design modifications that have the greatest impact on availability.
Input to AVPROG consists of system geometry (as defined on logic diagrams), mean time between failures, mean time to restore (MTTR),  and mean plant delta (MPD) for each logic block. MTTR is defined to  include access time, equipment preparation time, actual replace time, cleanup time, and time to restore the equipment and associated subsystem to normal operation. MPD is defined as additional time to  restore full production after theequipment in the logic block has been restored.
top ]
4. METHOD OF SOLUTION

AVPROG uses Monte Carlo simulations to estimate availability based on time histories of production developed from failure and restore data for each logic block (on the logic diagram  representing the system operation) and from specified additional restore time data for the failed system.
top ]
5. RESTRICTIONS ON THE COMPLEXITY OF THE PROBLEM:
top ]
6. TYPICAL RUNNING TIME

Execution time is dependent on the number of  blocks in the system, the number of batches, and the simulation time (years) for each batch. A problem, consisting of 49 blocks run at 10 batches for 5000 years each, produced an average of 13,446 failures  per batch with a standard deviation of 0.0000015 and took 2.75 minutes of CPU time. A much more complicated problem with 1048 blocks run for 10 batches for 500 years produced an average of 70,749 failures per batch with a standard deviation of 0.000014 and  required almost 129 minutes of CPU time.
NESC9700/01
NEA-DB executed the test case included in this package  on an IBM 3084 computer in one second of CPU time.
top ]
7. UNUSUAL FEATURES OF THE PROGRAM

AVPROG is unique in its ability to handle MPD in a parallel logic configuration. AVPROG can handle K-out-of-N parallel configurations where the components are not identical. It can also handle weighted and normalized parallel configurations. In considering parallel configurations in the logic  diagram, all components are assumed to be operating normally and not on standby. The output shows not only the system availability, but also the contribution of each block to system unavailability, ranked in order of size of the contribution.
top ]
8. RELATED AND AUXILIARY PROGRAMS

AVANAL can use the AVPROG input data to calculate an exact availability for the system. It is frequently used to determine an upper limit of system availability or find errors in input data before AVPROG is run.
top ]
9. STATUS
Package ID Status date Status
NESC9700/01 18-MAY-1987 Tested at NEADB
top ]
10. REFERENCES

- AVPROG, NESC No. 9700.3033B, AVPROG Tape Description,
  National Energy Software Center Note 87-86, August 28, 1987.
NESC9700/01, included references:
- AVPROG: Availability Analysis Procedures Manual.
  Y/PO-87-1  (August 1,1983)
- AVPROG: User's Manual.
  Y/PO-87-02, Rev. 1  (July 25, 1985)
top ]
11. MACHINE REQUIREMENTS:  Program size: 372K bytes.
NESC9700/01
To run the test case on an IBM 3084, 380K bytes of main storage are required.
top ]
12. PROGRAMMING LANGUAGE(S) USED
Package ID Computer language
NESC9700/01 FORTRAN-IV
top ]
13. OPERATING SYSTEM UNDER WHICH PROGRAM IS EXECUTED:  MVS/JES2.
NESC9700/01
MVS (IBM 3084).
top ]
14. OTHER PROGRAMMING OR OPERATING INFORMATION OR RESTRICTIONS

Two
subroutines are written in assembly language; they are IDAY, which returns the current date, and CORE, which is used to convert character data to integer data.
top ]
15. NAME AND ESTABLISHMENT OF AUTHORS

          P.W. McGrady and L.D. Reynolds
          Martin Marietta Energy Systems, Inc.,
          P.O. Box, P
          Oak Ridge, Tennessee 37831
top ]
16. MATERIAL AVAILABLE
NESC9700/01
File name File description Records
NESC9700_01.001 Information file 67
NESC9700_01.002 AVPROG source (FORTRAN) 2669
NESC9700_01.003 Subroutines (ASSEMBLER) 330
NESC9700_01.004 Input data for sample problem 34
NESC9700_01.005 Printed output of sample problem 138
top ]
17. CATEGORIES
  • P. General Mathematical and Computing System Routines

Keywords: Monte Carlo method, availability, production, simulation, systems analysis.