Supplier Evaluation using Analytic Hierarchy Process (AHP) Under Guidance of Prof. V.N.A. Naikan
By Subhani Shaik M.Tech 2nd Year, 05IM6013 Dept. of Industrial Engineering and Management. IIT Kharagpur.
Company Information Problem Definition Project Objective Project Scope Methodology Literature As-Is Process Map Data Collected Analysis Work Completed Future Plan of Work
Company Information Honeywell Technology Solutions Lab (HTSL) is an integral arm of Honeywell International.
HTSL provides value to Honeywell businesses through Product Solutions, New Product Introduction, Advanced Research & Technology and IT & Business Process Solutions.
HTSL supports and develops products in two key areas of Honeywell businesses. 1. Aerospace 2. Automation and Control Solutions.
Company Information… In Aerospace HTSL develop and support ‘safety critical aerospace systems’ like Flight Management Systems, Flight Panel Displays, and Engine Control Systems to its aerospace clients Airbus and Boeing. In Automation and Control Systems , HTSL products are Access systems , Burglar Alarms, Switches and Sensors. HTSL also develops necessary IT & Business Process Solutions for its products.
Problem Definition Honeywell Technology Solutions Lab (HTSL), Bangalore is a Product Development and Research Lab. For its new product development, the company is forecasting the demand for its customers. The designers are developing drawings for each part of a product using computer aided techniques like CAD. HTSL is sending its part drawings to its different suppliers for their prototype development. The suppliers are manufacturing prototypes as per the drawings provided. In Honeywell there is no proper system to evaluate the prototype suppliers based on their performance. HTSL needs proper system to evaluate its prototype suppliers.
Project Objective Evaluation of prototype suppliers based on their performance.
Project Scope The dissertation work covers the evaluation of prototype suppliers of mechanical components of Sensing & Control department.
Methodology
Find the current process of evaluation of suppliers
Identify the key supplier performance factors
Identify the methods to measure supplier performance factors
Data collection from the existing system
Analyze the data collected
Measure supplier performance factors
Identify best method to evaluate overall performance of suppliers
Find the overall performance of the suppliers
Ranking the suppliers based on their overall performance
Review supplier performance continuously
Literature Key Supplier Performance Factors Supplier Evaluation Methods Performance Factor - Quality Performance Factor - Delivery Performance Factor - Cost Performance Factor - Service
Literature : Key Supplier Performance Factors Quality Delivery Cost Service Supplier Management Capability Overall Personal Capabilities Financial Capability and Stability Information System Capability Environmental Regulation Compliance Supplier Purchasing Strategies and Policies Long Term Relationship Potential
Literature : Supplier Evaluation Methods Method
Reference
Quantitative / Qualitative Parameters
Advantages
Disadvantages
Categorical
Timmerman (1986)
-Quality -Delivery -Service -Price
-The evaluation process is clear and systematic -Inexpensive -Requires a minimum Performance data
-Attributes are Weighted equally -Subjective -Imprecise
Weighted Point
Timmerman (1986)
-Quality -Delivery -Service -Price
-Attributes weighted by Importance
-Subjective -Difficult to effectively consider qualitative criteria
Cost Ratio
Timmerman (1986)
-Quality -Delivery -Service -Price
-Subjectivity is reduced -Flexibility
-Complexity and requirement for a developed cost accounting system -Performance Measures (cost ratios) are artificially expressed in the same Units
Literature : Supplier Evaluation Methods… Method
Quantitative / Reference Qualitative Parameters
Total cost of ownership
Ellram (1995)
Analytic Hierarchy Process (AHP)
Advantages
Disadvantages
-Price -Quality Costs -Unreliable Delivery Costs -Transport Cost -Ordering Cost -Reception Cost -Inspection Cost
-Substantial Cost savings -Allows various purchasing policies to be compared with one another
-Complex
Nydick & Hill (1992)
-Quality -Delivery -Price -Service
-Simplicity -Captures both qualitative and quantitative criteria
-Inconsistency on the method
Principal Component Analysis
Petroni & Braglia (2000)
-Quality -Delivery -Price -Reliability
Considers simultaneously multiple inputs and outputs without priori assignment of weights
-Knowledge of advanced stati-stical method is required
Neural Networks
Wei (1997)
-Performance -Quality -Geography -Price
-Saves a lot of time and money of the system development
Lack of experts and Requires a software
Literature
: Performance Factor - Quality
Traditional Definition : Quality means fitness for use. Modern Definition
: Quality is inversely proportional to variability (Douglas C. Montgomery)
Quality can be measured in one of the following ways: 1. Defective Parts Per Million ( PPM) 2. Sigma Quality Level / Process Sigma 3. Process Capability 4. Rating Method
Literature : Performance Factor – Quality… PPM for Continuous Data : Let p= Prob. that produced items not meeting requirements p= P(XUSL) PPM =p * 1,000,000 PPM for Discrete Data : Total no of defective items PPM = ------------------------------ X 1,000,000 Total no of items produced
Literature : Performance Factor – Quality… Sigma Quality Level / Process Sigma Six Sigma Quality Level means “All parts or processes within six standard deviations on each side of mean are acceptable. “ For the given Six Sigma Quality Level: For Normal Distribution centered at Target (i.e. Mean = Target ) 99.9999998% parts are acceptable. i.e. Only 2 parts per billion(0.002 parts per million) are defective. Sigma Quality Level 6σ 5σ 4σ 3σ 2σ 1σ
Percent Accepted 99.9999998% 99.999943 99.9937 99.73 95.45 68.27
Defective PPM 0.002 0.57 63 2700 45500 317300
Note : 3.4 parts per million are defective if the mean is shifted by 1.5σ from Target for the given Six Sigma Quality Level.
Literature : Performance Factor – Quality… Process Capability : It is the capability of a process to produce a product relative to the stated tolerances. Cp and Cpk are two statistics used to define process capability. Cp compares the width of the data set variation to the width of the specification. Where as Cpk compares the width and centring of the data set to the specification target , upper and lower values. USL – LSL Process Capability Ratio Cp = ---------------
R-Bar Where σ = Estimated Sigma = ---------
6*σ
d2
min (USL-μ, μ-LSL) Process Capability Index Cpk = -------------------------3*σ
Literature : Performance Factor – Quality… Rating Method : In Rating Method, we are allocating weights to different sub factors of quality. • Incoming / Manufacturing Results (60%) Defective Parts Per Million (PPM) (or) Sigma Quality Level (or) Process Capability • Response to Quality Audit Observations
(20%)
• Response to Corrective Actions Requests (20%) Quality Rating = 0.6 * Incoming Result + 0.2 * Response to Quality Audit Observations + 0.2 * Response to Corrective Actions Requests
Literature : Performance Factor - Delivery Delivery can be measured in the following methods Defective Parts Per Million (PPM) Rating System Defective Parts Per Million (PPM) No of times delivery is not on time PPM = ------------------------------------------ X 1,000,000 Total No of times delivered Rating System (Delivery before / after delivery date) 1 More than One week
3 Up to One Week
5 3-4 days
7 1-2 days
9 On time
Literature : Performance Factor - Cost Cost factor in supplier evaluation process can be measured by Rating method as follows.
Cost : Sub-factors i.
Max Points
Cost Reduction Suggestions
5
ii. Net Cost Reduction Performance
5
iii. Cost Management Initiative
5
iv. Performance during product delivery process
5
Total Points
20
Literature : Performance Factor -Service Service factor in supplier evaluation process can be measured by Rating method as follows. Service : Sub-factors
Max Points
i. Proactive Communication
10
ii. Responsiveness
10
iii. Extraordinary Arrangements
5
iv. Accessible / Diligent
5
v. Flexibility
5
Total Points
35
As-Is Process Map : Current Inspection Process
As-Is Process Map : Current Supplier Evaluation Process
Data Collection Quality related data collected from Quality Control department . Prototype Samples Dimensions Response to Quality Audit Observations Response to Corrective Actions Requests. On time delivery information of suppliers was collected from Stores Department. Cost related data collected from Purchase Department. Cost Reduction Suggestions Net Cost Reduction Performance Cost Management Initiative Performance during product delivery process Service related data collected from Purchase Department. Proactive Communication Responsiveness Extraordinary Arrangements Accessible / Diligent Flexibility
Analysis Quality Defective Parts Per Million (PPM) (60%) Response to Quality Audit Observations (20%) Response to Corrective Actions Requests (20%) Delivery PPM Cost Cost Reduction Suggestions Net Cost Reduction Performance Cost Management Initiative Performance during product delivery process Service Proactive Communication Responsiveness Extraordinary Arrangements Accessible / Diligent Flexibility Analytic Hierarchy Process
(10 Points) (10 Points) (5 Points) (5 Points) (5 Points)
(5 (5 (5 (5
Points) Points) Points) Points)
Analysis : Quality - How to find PPM ?
Prototype Sample Data
Analytic Hierarchy Process (AHP) Analytic Hierarchy Process (AHP) is a multi criteria decision technique that can combine qualitative and quantitative factors for prioritizing, ranking and evaluating alternatives. AHP reduces complex decisions to a series of one-on-one then synthesizes the results.
comparisons,
AHP Measurement Scale Comparing objective i and objective j We set aii = 1. if we set aij = k, then aji =1/k Verbal Judgment of Preference
Numerical Rating
Objectives i and j are of equal importance
1
Objective i is weakly more important than j
3
Objective i is strongly more important than j
5
Objective i is very strongly more important than j
7
Objective i is absolutely more important than j
9
2,4,6,8 are Intermediate Values provided additional level of discrimination
Analytic Hierarchy Process (AHP) Steps 1.Specify the set of criteria for evaluating the supplier’s proposals. 2.Obtain the pair wise comparisons of the relative importance of the criteria in achieving the goal, and compute weights of the criteria based on this information. 3.Obtain measures that describe the extent to which each supplier achieves the criteria. 4.Using the information in step 3, obtain the pair wise comparisons of the relative importance of the suppliers with respect to the criteria, and compute the corresponding weights. 5.Using the results of steps 2 and 4, compute the priorities of each supplier in achieving the goal of the hierarchy.
Work Completed
Theoretical Framework Documenting Existing Process Data Collection Analysis AHP to Evaluate Supplier Overall Scores Ranking of Suppliers
Work to be done Consistency Checking of AHP
Future plan of work This dissertation work can be expanded to other sectors like manufacturing Additional Criteria can be considered for supplier evaluation Compare supplier evaluation issues with ISO 9000 standards Study of other supplier evaluation methods like Principal Component Analysis (PCA) , Neural Networks etc.
References DOUGLAS C MONTGOMERY (2004) Introduction to Statistical Quality Control - Fourth Edition John Wiley & Sons, Inc ELLRAM (1995) “Total Cost of Ownership: An Analysis Approach for Purchasing”, International Journal of Physical Distribution and Logistics, pp. 163-184. EUGENE L GRANT AND RICHARD S LEAVENWORTH Statistical Quality Control -Seventh Edition. NYDICK AND HILL (1992) “Using the Analytic Hierarchy Process to Structure the Supplier Selection Procedure”, International Journal of Purchasing and Materials Management, 1992, pp.31-36. PETRONI AND BRAGLIA (2000) “Vendor Selection Using Principal Component Analysis”, The Journal of Supply Chain Management: A Global Review of Purchasing and Supply, pp. 63-69. RICHARD A JOHNSON AND DEAN W.WICHERN Applied Multivariate Statistical Analysis - Fifth Edition
References… SAATY (1980) The Analytic Hierarchy Process. NY: McGraw-Hill SIMPSON, SIGUAW AND WHITE (2002) “Measuring the Performance of Suppliers: An Analysis of Evaluation Processes”, The Journal of Supply Chain Management, pp. 29-41. TIMMERMAN(1986) “An Approach to Vendor Performance Evaluation”, The Journal of Supply Chain Management, pp. 2-8. WEBER, CURRENT AND BENTON (1991) “Vendor Selection criteria and methods”, European Journal of Operation Research, pp. 2-18. WEI, JINLONG AND ZHICHENG (1997) “A Supplier Selecting System using a Neural Network”, IEEE International Conference on Intelligent Processing Systems, pp.468-471.
Thank You