Scribd
Upload
75 views

Ans: - Continous and Discrete Simulation Model

The document discusses continuous and discrete simulation models. Continuous models have smoothly changing states while discrete models have discontinuous state changes at distinct event times. Random number generation techniques for simulations are also described, including linear congruential generators and tests for uniformity and independence of random numbers. Data collection methods and determining the appropriate length and number of simulation runs is also summarized.

Uploaded by

Anitesh Saunders
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
75 views

Ans: - Continous and Discrete Simulation Model

The document discusses continuous and discrete simulation models. Continuous models have smoothly changing states while discrete models have discontinuous state changes at distinct event times. Random number generation techniques for simulations are also described, including linear congruential generators and tests for uniformity and independence of random numbers. Data collection methods and determining the appropriate length and number of simulation runs is also summarized.

Uploaded by

Anitesh Saunders
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 6

Assignment of Simulation and Modelling. 1.Explain the computation technique for Continous model and Discrete model?

Ans:->Continous and Discrete Simulation Model:->Systems or the model in which the changes are Predominantly smooth are called Continous simulation Model.ex:-An Aircraft System. While in discrete simulation System, changes are predominantly discontinuous.Arrival and Departures of customers occoured at a distinct points of time this is what we refer to as a events.When the number of events are finite ,we call the simulation Discrete Events. Events.Examples:- Factory Systems.In some System the state changes all the time, not just the time of some discrete Events.The water level in a reservoir with given in and outflows may change all the time .in such cases,Continous Simulation is more appropriate although discrete event simulation can serve an approximation .few System are completely continuous or Discrete .e.g. :->an Aircraft System can be regarded as an Discrete System. In Aircraft we may take Discrete adjustments to trim as altitude changes while in factory Systems machinery is continuous activity and start and finish of a job are as discrete changes.But in most of the System one type of change continuous or Discrete predominates, so that we can categorized our System as being Continous or discrete. 2.Sampled Data simulation Model:-> There are also some Systems where information about them is only available at discrete points in time but system are intrinsically continous. 3.Explain the method of generating Random numbers and also Explain how to test Random ? Ans:-> Random numbers is very important in different areas of computer Applications.we have seen that the demand of a particular product can go up,down or stay at the same level at any particular day.Let the probabilities of these three items are 0.3,0.5,0.2 respectively.Then we need a Random Selector which reproduces outcome in the longrun with relative frequency 0.3,0.5 & 0.2.we are having number of techniques to generate pseudo random numbers . These techniques are usually based on some recurrences relation.These techniques are 1.Linear congruential Generators. 2.Mixed Generators. 3.Multiplicative generators. 4.Combined Linear Congruential generators. 5.Tausurtho & Related Generators.
1

5.5.1 Techniques for generating Random Numbers. 1.Techniques for generating Random numbers: 1.Linear Congruential Generators(LCG) Zi=(aZi-1+c)(mod m) m-Modulas.
a-Multiplier.

c-Increment. Z.-Seed or starting value. Ui=zi/m.


#TESTING OF RANDOM NUMBERS: ->The basic properties that we have to need to test Random

numbers is
1.Uniformity. 2.Independence. 1.Uniformity test:->Uniformity test counts how often numbers in a given range occour in the sequence to ensure that the numbers are uniformly distributed.there should be no number that can occour more Frequently.Two methods are given to perform Uniformly test. 1.KOLMOGOROV SMIRNOV TEST. 2.CHI-SQUARE TEST. 1.kolmogorov smirnov test:->In this test we compare the continous cdf(continous distributed function) ,F(X),of the uniform distributed to the empirical cdf,En(x),of the sample of n distribution.

2.kolmogorov smirnov test is based on the largest absolute deviation between f(x) and En(x) over the range of the random variable.it is based on the statistics. D=max/F(X)-En(x)/ D is known as sample distribution.
2

3.CHI SQUARE TEST :- chi2 =

(Oi-ei)2/ei;

n=number of classes. oi=number of times the observed data falls in each class I for i=1,2,3,..n; ei=N/n; N=Total number of observations. 5.Define the terms: 1.length of the simulation runs. 2.Simulation data collection and Reduction. 3.Simulation Starting condition and Experimental Layout. 1.The two widely used method for experimentation on simulation models are method of bath means, and independent replications. after deciding what method is more suitable to apply ,the main question is determination of number of runs.that is at the planning stage of simulation investigation of the question of the number of simulation runs(n) is critical . 2.simulation data collection and reduction:->when the needed statsistics for the number of simulation run calculation is not available from existing data base ,a pilot simulation is needed.for large pilot simulation run,say over 30 ,the simplest number of runs determinate is [(za/2)*2]/d*d;

7.runs size with Acceptable Absolute Precision 1.pilots runs size: 2.Current Estimate: 3.Current Variance Estimate: 4:Acceptable Significant Level: 5.Acceptable Absolute Error: 6.The required runs size is:
3.sources of data:-> 1.Computer files. 2.Database. 3.SpreadSheets. 4.RawData. 5.Video Records. 6.Hard copy files and Documents. 7.Interviews. 8.Video. 9.Observation of running systems. 10.Interviews.

3.1 WHEN AND HOW MUCH DATA TO COLLECT.


3.2 To help design model. 3.3 interviews. 3.4 Observations. 3.5 MinimumNecessary. 3.6 To obtain i/0 parametres.

4. Data Collection report. 4.1 A description of completed Collection process that includes 4.2 Restatement of plan and any changes to original plan. 4.3 Result of Statistical analysis of data. 4.4 Graphs , average , max and min. 4.5 Summary of results. 4.6Relations of results to client Objectives. 4.7Recommendations. 4.8 Diagrams as necessary.

SUBMITTEDTO: Mr.PRADEEP SAHINI. DEPARTMENT OF CSE & I.T.(S.H.I.A.T.S) SUBMITTED BY:- Anitesh John Saunders 08BTCSE004. SEM :-8TH.

You might also like