• 沒有找到結果。

Chapter 2 LITERATURE REVIEW

2.2 Performance Evaluation

2.2.1 Definition of Performance Evaluation

The significance of performance evaluation is that different scholars have different explanations whose views can be summarized into the following several points:

1. Performance evaluation can be an objective and the quantitative data collection and analysis, in terms of a digital or percentage form of its efficiency and effectiveness (Kantor, 1984).

2. Performance evaluation is a management tool, under a meaningful organization assessment, can get a tangible performance data (Cronin & Mary, 1985).

3. Performance evaluation is a kind of investigation and analysis of the existing situation, is also a kind to change the status quo and promote efficiency of the process and management tools, to make the investigation and analysis of the data to be useful and in accordance with the purpose of the researchers, be must select the appropriate evaluation method (McClure, 1986).

Chiao-Pin Bao at the performance evaluation comprehensive application (2008), present the significance of performance evaluation; to measure, evaluation organization and members, during a time of job performance.

There are many different methods of performance evaluation method: The Ratio Analysis (Sun, 2004), the Regression Analysis (Sun, 2004), the Analytic Hierarchy Process (AHP) (Saaty, 1977, 1980), Grey Relation Analysis, GRA) (Ju-Long Deng, 2003), Balanced Scorecard (BSC) (Cui-Shuang Gao, 2000), Stochastic Frontier Approach (SFA) ( Aigner et al.,1977) etc.

2.2.2 Techniques for Performance Evaluation Method

This section will assess the advantages and disadvantages of the performance evaluation methods in the following table.

Table 2-1: The comparison of performance evaluation methods

Method Advantage Disadvantage

Ratio analysis method

1. Ease of use, does not need too much theoretical foundation.

2. Through the characteristics of the results can be clearly assess performance, to distinguish between good or very bad efficiency.

1. Only for a single target, Can't represent all the working efficiency.

2. Cannot provide inefficient unit amplitude of improvement project.

3. Relating to input and output must have the same unit of measure 4. The results of evaluation of single ratio value alone.

5. Different unit performance indicators, more difficult to judge.

The Regression Analysis Method

1. Will be multiple inputs as independent variable and output as the dependent variable, in order to predict life force.

2. The express the relationship between input and output functions, objective analysis.

3. With statistical analysis theory, the basis of analysis results of science and technology.

4. Under the condition of limited samples limit will not be

inefficient unit as efficient units.

1. Can set one dependent variable, unable to deal with multiple output problems.

2. In the assessment units sample is small, can't find out the most efficiency of the unit.

3. Between the independent variable and dependent variable with linear function relation, otherwise the efficiency value will not be accurate.

4. Not exactly pointed out that organization which is efficient.

Table 2-2: The comparison of performance evaluation methods ( cont )

Method Advantage Disadvantage

Production frontier method

1. Using simple statistical test and evaluation results more objective

2. Less restrictive conditions, the mathematical structure is simple and economic implication is clear.

1. All input and output items should be quantifiable

2. Need to set production function type, and only a single output.

3. Residuals shall be assumed as normal distribution, otherwise unable to take the production function.

BSC (Balanced score card)

1. Link purpose, assessment, target and action of the four dimensions as the foundation of evaluation.

2. Organization operation results will be used for internal

communication, learning tools, rather than the only exception 1.

3. The output indexes are not performed within culvert management control purposes.

1. Grade appraisal objective and fair enough.

2. Only one of the indicators to assess the efficiency of operation, cannot represent the overall efficiency.

3. Performance evaluation indicators must be through the experts to give score, not enough for objective and impartial.

Grey Relation Analysis, GRA)

1. Can deal with uncertain or not complete information.

2. Not strictly required the number of sample, don't need big data.

1. Data must conform to the principle of comparability, and must be

normalized processing.

2. Advantages disadvantages order can only do the comparison, limited decision-making information.

Stochastic Frontier Approach, SFA

1. Can use the set at random item to deal with the effects of

interference factors such as statistical error.

2. Evaluation can more close to the actual production situation.

1. Need more observation points, the efficiency of the parameter values have higher accuracy

Table 2-3: The comparison of performance evaluation methods ( cont )

Source: Reference (Sun, 2004) and this study summarized

Method Advantage Disadvantage

Analytic Hierarchy Process, AHP

1. Suitable for most unstructured problems.

2. Can simplify the problem reduce the complexity of the judgment.

1. When evaluating index is large, the number of the require cost have changed much.

Data Envelopment Analysis, DEA

1. Can simultaneously measure the efficiency of dealing with multiple inputs and multiple outputs.

2. Prior assumptions form the relationship between the

production function, the parameter estimation problem can be

avoided.

3. Not affect the enterprise's subjective consciousness. Input and output data must be very accurate, efficient frontier is efficient commitment and the right output value items produced by the mathematical programming model.

4. It is a comprehensive index evaluation result, easy to do comparative efficiency.

5. Can provide use of DMU and improve the policy.

1. Must deal with the huge input and output data.

2. Homogeneity makes sense when measuring of the effect is high.

3. When numerical input and output items are negative, it will be unable to process.

相關文件