W10._Norbert Eze_ Selecting Suppliers by Using Compensatory Models

W10_Norbert Eze_ Selecting Suppliers by Using Compensatory Models

By norberteze on December 31, 2012

I.      PROBLEM RECOGNITION, DEFINITION AND EVALUATION

The cfh has instructed to apply COMPENSATORY models to the same case study…”A supply chain manager of NETCO has instructed me to carry out selection of a best supplier to purchase some materials from” and what do compensatory models provide us with that non-compensatory models do not. The advantages and weakness in each these model. Table…1 shows the desired attributes to assist in the alternatives evaluation of the suppliers. The compensatory model is of two types as below:

a)    Non-dimensional Scaling

b)    (b)Additive Weighting Technique

Compensatory Models are multi-attribute decision making tools used to collapse information into a single dimension and can make complex problems computationally tractable in a particular manner. It is called compensatory because the changes in the values of a particular attribute can offset or trade off against opposing changes in another attribute.

II.    DEVELOPMENT OF FEASIBILITY ALTERNATIVES

To develop the feasibility alternatives Table…1, is the starting point of attributes and alternatives filled with objective and subjective data relating to differences among the attributes. I am maintaining the table used in W9 as below:

                                                                                                                                        

Non-dimensional Scaling:

A popular way to standardize attribute values is to convert them to non-dimensional form. The non-dimensional values should have a common range, such as 0 to 1 or 0 to 100. Without this constraint, the dimensionless attributes should follow the same trend with respect to desirability; the most preferred values should be either all small or all large. This is necessary in order to have a believable overall scale for selecting the best alternative.

The procedure for converting the original data in Table…1 for particular attribute to its dimensionless rating is

Rating   =        Worst outcome – Outcome being made dimensionless              1.1

                                    Worst outcome - Best outcome

Equation 1.1 applies when large numerical values, such as cost or distance, are considered to be undesirable. When large numerical values are considered to desirable, however, the relationship for converting original data to their dimensionless ratings is shown in Equation 1.2

Rating   =        Worst outcome – Outcome being made dimensionless              1.2

                                    Worst outcome - Best outcome

As shown in Table…6, the preceding constraints may require that different procedures be used to remove the dimension from each attribute. Rating Procedure that is to say Equation 1.1 and 1.2 are applied below:

Non-dimensional Data for choice of Supplier spread in Table…7 as required.

Using spreadsheet solution for Non-compensatory scaling: See table…8 below.

The dimensionless rating for each alternative/attribute combination is determined by comparing the difference between the alternative under consideration with the lowest scoring alternative. This difference is then divided by the range between the best and worst scoring alternatives to arrive at the non-dimensional value. The total score for each alternative is found by summing these values.

See the formulas in the highlighted cells in Table…9  below derived from Table…8

Additive Weighting Technique

Additive weighting provides for the direct use of non-dimensional attributes and the result ordinal ranking. The procedure involves developing weights from attributes (based on ordinal rankings) that can be multiplied by the appropriate non-dimensional attribute values to produce a partial contribution to the overall score for a particular alternative.

 

Additive weighting is probably the most popular single dimensional method because it includes both the performance ratings and the importance weights of each attribute when evaluating alternatives. It produces recommendations that tend to agree with the intuitive feel of the decision maker concerning the best alternative. The biggest advantage is that removing the dimension from data and the weighting attributes are separated into two steps. This reduces confusion and allows precise definition.

See the application of Additive Weighting Technique spreadsheet in Table…10 of Selection of Supplier.

See the exact result of application of Additive Weighting Technique in Table…11 of Selection of Supplier.

 

III.   DEVELOPMENT OF OUTCOMES FOR EACH ALTERNATIVE

Applying Compensatory Models using Non-dimensional Scaling to determine outcomes for each alternatives using Table..8

·      Supplier A scored the highest mark based on the fact that attributes have Experience, Quality, Skill, Reputation and Capacity 1.00, 0.33, 1.00, 0.00 and 1.00 respectively.

·      Supplier B ranked second based on the fact that attributes have Experience, Quality, Skill, Reputation and Capacity 0.50, 1.00, 1.00, 0.50 and 0.00 respectively.

·      Supplier C ranked third based on the fact that attributes have Experience, Quality, Skill, Reputation and Capacity 0.25, 0.67, 0.00, 1.00 and 1.00 respectively.

·      Supplier D is the last based on the fact that attributes have Experience, Quality, Skill, Reputation and Capacity 1.00, 0.33, 1.00, 0.00 and 1.00 respectively.

Applying Compensatory Models using Additive Weighting Technique to determine outcomes for each alternatives using Table…11

·      Supplier A scored 0.59

·      Supplier B scored 0.67

·      Supplier C scored 0.60

·      Supplier D scored 0.06

IV.  SELECTION OF A CRITERION (OR CRITERIA)

The Additive Weighting Technique is friendly user compare to non-compensatory scaling. Although, the calculation in the non-compensatory scaling seen to have some issues which will be investigated latter in my next blog. Additive Weighting Technique gives credit to Supplier B and while Compensatory Models gives credit to Supplier A. In Additive Weighting Technique has highest score of 0.67 and Non-compensatory Scaling has highest of 3.33. More investigation will be carried out to know the reasons behind this inequality.

V.    ANALYSIS AND COMPARISON OF THE ALTERNATIVES

Using Additive Weighting Technique, the best supplier is Supplier B that scored 0.67. Experience, Quality, Skill, Reputation and Capacity score 0.17, 0.44, 0.00, 0.06 and 0.00 respectively. Find more information Table…11

VI.  SELECTION OF PREFERRED ALTERNATIVE

Preferred alternative is Supplier B. The best attribute scored Supplier B best and best attribute is Quality of Products which is ranked 4. Selection of Supplier B has go through many tests and has been deemed it right to be the best. So the contract will be given to Supplier B.

VII. PERFORMANCE MONITORING AND POST-EVALUATION OF RESULT

The application of Non-compensatory Scaling and Additive Weighting Technique should be exploited in the selecting, monitoring and post-evaluation of result of Supplier.

References:

1.      Purdue OWL APA style. (2011). APA formatting and style guide, pg (10). Retrieved from http://owl.english.purdue.edu/owl/resource/560/19

2.      Sullivan, W. G., Wicks, E.M., & Koelling, C.P. (2012). Engineering economy, (15th ed.)  (Chapter 1) (pp. 23-37)  (Chapter 14) (pp. 573-591)

3.      Howard Technology Middle East (2012). Operating Cost Estimating and Financial Analysis, ( 2012) (Selection 7) (pp. 1-7)

4.      AACE International. (2012). Skill & Knowledge of Cost Engineering, (5th ed.)  (Selection 5) (Chapter 23) (pp. 23.1-23.10) (Chapter 24) (pp. 24.1-24.5)

5.      Canada, et al., Captital Investment Decision  Analysis for Engineering and Management, 3rd ed. (Upper Saddle River, NJ: Prentice Hall, 2005)