The Evaluation System Of Human Work Potential Based On Grey System Theory

As development of society, Human resources is becoming more andmore important in many companies, so there are many chances that thesystem of job performance is used to judge the performance of employee.But, it always pays more attention onthe former performance of work. Thiswill bring much chanciness to the final decision-making. So, this papergives a predictive method based on the prediction model of Grey Systemtheory, which can predict the ability to work. Heads of department can getcomprehensive knowledge about the ability of employees to make asensibledecision.The Grey System is an uncertainty theory that studies a little data. Itscontent of the study includes data processing, phenomenon analyzing,establishment of model, trend prediction, decision-making of the thing,controllingofthe system and evaluationof thestate. It canprovide aproofforcreatinga relationofmanandnatureoreconomyandresources, offer asupportforresolvingoptimaluseofresourcewithlimitedinformation.Prediction pays attention on the activity of the changes in law of thedevelopment of the society and technology. It predicts the trend of futurebythe data accumulated. There are two stages in the prediction. One is thesummarized process. It starts at identifying target and collecting the datathat has a relationship with the prediction target, then uses model to findout the evolution disciplinarian after processing data. The other is theinference process. It uses the evolution disciplinarian to predict theperformanceofthetargetbytheknowledgeoftheconditionsofthefuture.Grey prediction finds out and holds the disciplinarian of the systemdevelopment and predicts the intending state after processing the primaldata and establishingthe greymodel. All of the greymodel can be used as prediction model. So, when facing a problem, we should choose aappropriate prediction model by fully analyzing. One model's precisionand the prediction data only can be accepted after checking up. Of course,themodelthatisonlythroughtestingcanbeusedtopredictdata.GM model is different from other models that use primal data toestablish the differential equation. It establishes the differential equationafter processing the primal data. The primal data may show the discretestate because of the system is polluted bythe dirtydata. The system that isestablishedbythesedirtydatanamedgreymodel,abidingbythefollowingconditions.1. The grey system theory regards the random variables as greyvariableswithinacertainrange.2. The essential of GM model is generation of the data series, becauseitmakesthediscretedatadisciplinary.3. The greysystem defines the time range of data series, then, definingtheconsistenceofinformation.4. The grey system enhances the accuracy by adopting different thewayofgeneration.5. The GM model adopts three methods to check up the precision,respectively, residual test, correlation test and after the post-mortemexaminationtest.6. The data from GM model can be used to predict after inversegeneration.The GM model is used in this system.. But when the primal data isused in predicted, it must be processed in the following, because they aredirtydataduringbeingcollected.1.Calculatetheminimumnamedx(min)intheprimaldata.2. Make sure the relationship between the total number of primal dataand the number of data that join in prediction. After processing theirformislikethis: x⁰ = {[ x⁰ ( n - m + 1) - x (min) + 1],[ x⁰ ( n - m + 2) - x (min) + 1], K[ x⁰( n ) - x(min) +1]}Thenisthetotalnumberoftheprimaldataandthemisthenumberofdata thatjoininprediction.3.Primaldatawereaccumulatedgeneratingandaveraged.When predicting the ability of work, the different models should beused in different working stages. Facing the freshman, the GM(1,1) can beused to predict their ability of work because of their examination resultsaccording with the exponential growth trend. But we cannot use this modelto predict the old staff, because there are some fluctuant points in theirexamination results. So the GM(1,1) model must be improved on forpredictingmore accurately. Weuse Fouriertransform intheGM(1,1)modelfor resolving this problem and in order to improve the precision of theprediction. At last, this model is used in the evaluation system of humanworkpotential.