How to Select Suppliers through the AHP Model?

Christine Tsui Fashion Community

How to Select Suppliers through the AHP Model?

Kelly-Hangzhou- Vice admin of the fifth Christine Tsui Fashion community @ wkl2021

I used to repair cultural relics at China Silk Museum and have later become a designer brand merchandising buyer later. Now I am a textile purchasing buyer for furnishing projects. I usually like to read fashion and design magazines. In addition, I also like to visit major furnishing exhibitions and interesting furnishing stores, and I pay attention to the beautiful designs in life.

The Scenarios and Importance of Selecting Suppliers

“Supplier A wants to do business with us.”

The purchasing director was familiar with that supplier and said, “They can’t do it”

“He only knows this supplier on hand.”

“Isn’t there a supplier in Suzhou?”

“The price in Suzhou is too high.”

“What about Ningbo?”

“Poor quality.”

“What about Shanghai?”

“They don’t want to supply us.”

Do you find this conversation familiar? I believe those who work in fashion and clothing production industry have all encountered similar problems in supplier selection. Regardless of whether it is the one who selects, or the one who is selected, both parties need to know themselves and their counterpart very well to be able to match up accurately. Traditionally, supplier selection has been based on quantity and costs. Most companies have dedicated purchasing departments and archives of cooperating suppliers to share. Sometimes, even in each department, purchasing staff and even non-purchasing staff have their own cooperating suppliers. The company, through supplier market analysis in advance and consideration of many factors such as price, quality, delivery period, MOQ, etc., evaluates the strength and superior resource of each supplier and makes comparison of compares based on their compatibility, ultimately choosing the suitable supplier. In the fast-changing fashion market, flexibility and adaptability are the keys to survival. Therefore, supply chain strategies like rapid response are very important in the fashion and textile industry. Therefore, selecting a supplier to support this strategy is one of the most critical activities.

The traditional model is not the predecessor of the AHP model, but the study of this model gives the industry an example of a solution to one-stop hierarchical analysis process. The process takes into account operational performance (such as flexibility, cost, and delivery) to support the supply chain strategy. The system is implemented with the help of commercial software package experts’ selections. Based on the results of the hierarchical analysis model, the benefits of implementing fast response in the supply chain can also be testified. We will discuss this in detail below.

Variables and Criteria for Supplier Selection

Before getting to know the AHP model, we must first understand the variables and criteria for supplier selection that it involves. In recent years, with rising production costs and strict environmental regulations, more and more factories have moved production processes to inland China and other Asian countries, such as India, Vietnam and Thailand. Manufacturers in these countries need to understand what their retailers and customers need in order to compete with competitors from other countries. It also means that they need to know how customers select suppliers. By paying attention to these variables and standards, manufacturers can continuously improve their operation of supply chain.

In the fashion industry, the most basic variables for supplier selection are delivery date, quality, and price. The weight of each variable often depends on the buyer’s preference. If the volume is not large in the beginning, the price is usually the second concern. In addition, there are credibility, product style, qualifications and other cooperation terms. Among these, credibility refers to the need to consider the company’s background and business structure in advance to see if it is reliable, In addition the account period and payment method will also be taken into account. Credit is important to both parties. Some clothing companies have problems with the capital chain, so there may be defaulted payments to suppliers, and shopping malls sometimes also owe clothing companies money. Many suppliers have solid financial strength–if a clothing company owes money, they can only be the ones to pay factories and fabric suppliers first. But this is not a good way for long-term cooperation. For global business, credibility and background checks are even more important. Many companies will also buy credit insurance for protection. One of the services of this insurance is called credit survey, which is used to judge whether the customer is worthy of cooperation in terms of credibility. Credit Guarantee is supported by the Chinese government and belongs to the Ministry of Commerce. Although these consultations and related insurance require payment, they are more comprehensive and convenient than self-investigation. At present, the four state-owned banks also provide corresponding export insurance, but Credit Guarantee is more professional in terms of domestic trade. In addition, Tianyancha also has a relatively comprehensive collection of information for domestic companies. Why do we need to collect information? Because supplier selection is also essentially the screening of information, so accurate information sources are an important prerequisite.

Therefore, how to get official and correct information, such as that of the industry and suppliers, etc.? Generally, there are several ways:

  1. Websites of competent government departments, such as the National Bureau of Statistics;
  2. Industry associations, such as China Textile, etc .;
  3. Industry websites;
  4. Consulting companies and think tanks;
  5. Specialized data consulting service agencies, such as Credit Insurance,,, etc.;
  6. Other public data sources;
  7. Exhibitions, such as Shanghai Fashion Week, Fashion Shows, Import and Export Trade Fair;
  8. Actual samples.

In addition to credit, sometimes typical problems encountered by clothing manufacturers are the long distance between customers and suppliers, long production lead times, and seasonal product forecast errors. Fashion is characterized by a short product life cycle, so marketing time is a very important aspect of the industry.

For example, countries close to the US market are unlikely to be affected by competition from India and China. As a result, Mexico, the Caribbean, Eastern Europe and North Africa might remain important exporters of the United States and maintain their market share easily. On the other hand, Chinese suppliers are easily affected by factories in Southeast Asia.

For these reasons, manufacturers in Asian countries have to improve the efficiency of their supply chains and promote their competitiveness. For example, they apply the Quick Response (QR) philosophy. Flexible adaptation is the key to survival in future competition. In view of the above situation, whether it is for the buyer or the seller, the importance of supplier selecting is fathomable. The following table basically covers factors for consideration when selecting suppliers in the clothing industry.

图片 12031png副本

Based on this table, we can determine the variables or criteria for selecting the best supplier, including operating costs: flexibility (long distances, long production lead times, and seasonal product forecast errors), costs (time, money, Personnel), others (availability of raw materials, value-added costs, innovation capabilities, environmental planning issues, and environmental certifications), etc.; based on these variables, we can develop supplier selection models.

Management and Implementation of Quick Response in Supply Chain and Its Importance

Quick Response applies to activities throughout the supply chain-from producers to retailers. The aim is to reduce the time span from the stages of textile design to consumer purchase. Quick Response (QR) is one of the effective ways to improve the efficiency and competitiveness of the supply chain. In the face of a multi-variety and small-batch buyer market, companies do not reserve “products”, but rather, prepare various “elements”. When there are demands from users, they can extract “elements” as quick as possible and “assemble” in time, providing required services or products.

QR has basically become an important method in the fashion market to respond to the growing demand for diversity and long-term total lead times in global supply chains. Quick response requires both retail operators and textile manufacturers to cooperate, share information, and supplement products with high-frequency and small-quantity distribution to shorten delivery cycles, reduce inventory, and improve customer service and corporate competitiveness. Especially in the fast fashion industry, QR has a great impact on the competitiveness of enterprise.

How practical is QR in the actual production and supply process? QR is a big challenge for producers and suppliers, because it requires forecasting in advance, aligning production plan and so on. To achieve this, it is necessary to share information between suppliers and retailers to jointly cope with the changing market. The traditional production method is usually difficult to adapt to this pace. For a clothing supply chain, a clothing manufacturer is the core enterprise. Its upstream is fabric and trimmings supplier, and its downstream is clothing seller and customer. Garment manufacturers should actively seek and discover strategic partners and establish division of labor and collaborative relationship with them. They should be aware of their core position in the entire supply chain and build a chain network structure with their suppliers, suppliers of their suppliers and all forward relationships, as well as their users, users of their users, and every backward relationship. On top of that, establish a strategic partnership with them, refine the operation process, and reorganize the process to put core energies on the design of clothing and the application of new technologies, thereby improving the overall network chain’s responsiveness to multiple changes in the market. On the one hand, it is a test to the accuracy of forecasting and planning, and on the other hand, it is a trial of the supply chain’s capability of coordination. Nowadays, some fast-fashion brands have encountered one problem or another—this is because the market is changing, but the importance of the supply chain will remain. A very important concept of QR is to shorten the lead time of the supply chain and reduce inventory. Without QR, there would be too much (and expensive) inventory throughout the supply chain, resulting in discounted sales, or another extremity– the chance of stockouts would increase due to poor forecasts.


Modeling Analysis Based on Variables

What can the AHP model do about this? Selecting suppliers through AHP can achieve quick response to a greatest extent. “Analytic Hierarchy Process” (AHP) is a simple, flexible and practical multi-criteria decision-making method for quantitative analysis of qualitative problems. It features dividing various factors in complex problems into ordered and interconnected levels to make them organized; Based on the subjective judgment structure of a certain objective reality (mainly a pairwise comparison), it combines experts’ opinion and analysts’ objective judgment results in a direct and effective way, quantitatively describing the importance of the pairwise comparison of same-level elements.

Here is an example of AHP:



The figure above illustrates the procedure from “target” to the step of “information selection and comparison” to the final ” scheme selection”. The specific process is as follows:

When using AHP to model decision-related issues, the first step is to build a hierarchy. This issue is broken down into a systematic hierarchy, which is similar to a decision tree. The model has designed several layers of decision-making activities. The goal of the model is to determine which supplier is the best among the alternatives.

The goal is placed at the top of the hierarchy. The hierarchy goes down from the more general criterion at the second level to the sub-criterion. The four fully considered alternatives are at the bottom of the hierarchy.

Looking at it this way is a bit boring. Let’s take an example: there are four suppliers –A, B, C and D, and the goal is to choose the best supplier. Standard variables include delivery, quality, supply assurance, flexibility, cost, reliability of organizational strategy issues, perceived risk, technical logic issues, and environmental issues. Now we score all these aspects of each supplier, and built a model as shown in the figure:


Based on the current market and your situation, formulate your own criteria for selecting suppliers. Then I want to introduce something that seems a bit dull.


In this example, numerical comparison (matrix) is applied. If the numbers in the box are black, the criteria on the left is more important than that on the top. On the other hand, if the numbers in the box are red, the criteria on the top are relatively more important than that on the left. Figure 3 shows that while moderate delivery (value 3) is more important than “strong”quality ( value 5), flexibility and “very strong”cost  (value 7), delivery is not as important as guaranteed supply (value 3). Carrying on with this process, moderate quality (value of 3) is more important than flexibility, and  “strong” quality (value 5) is more important than cost and supply assurance (value 3), while a “very strong” supply assurance (value 7) is more important than flexibility, and “extreme” supply assurance (value 9) is more important than cost. In the end, moderate flexibility (value 3) is more important than cost.

Certainly, this requires the support of a computer system. Then you need to compare the priorities of each major and minor criteria. For example, highest priority of the total order lead time is 0.637, followed by the geographical location of 0.258. This indicates the overall weight information for each element in the analytic model.


In summary, after the model is established, pairwise comparisons are systematically performed to encompass all combinations of standard and substandard relationships and they form a matrix in each group on the hierarchy. For example, if there are four elements in one group, it will form a 4 × 4 matrix. Based on the collected data and information, via the nine-element ratio methodology, it carries out the comparison process according to the relative importance of the criteria and child criteria against the neighboring parent element.

In the end, the alternatives are compared by assigning corresponding values ​​based on the relative importance of the alternatives under each sub-criterion in the decision-making hierarchy. Then, the priorities of the analytic hierarchy process were calculated through the selection of commercial software experts. For the accuracy of the application, additional tests were conducted to reflect the consistency of the analytic hierarchy. Consistency measures can be used to filter the inconsistency in responses that are calculable through professional selection software. The standard relative weights are estimated at each level of the hierarchy. The matrices will get the relative weights of the elements in each matrix through a series of calculations. Relative weights are used to identify key factors. Criteria with higher relative weight are more important. After that, a series of sensitivity analyses are conducted to study the impact of changing standard priorities on the ranking of the alternates.

Another simple example can help us think further. For instance, if someone is going to buy a refrigerator, after knowing about 6 different types of refrigerators in the market, when deciding which one to buy, he usually does not compare directly, since there are many incomparable factors, Instead, he selects some intermediate indicators for investigation such as refrigerator capacity, cooling level, price, model, power consumption, external reputation, after-sales service, etc.

After this, he considers the pros and cons of various of refrigerators under the above-mentioned intermediate standards. With this ranking, a purchasing decision is finally made. When making decisions, since the order of each intermediate criteria is generally different with the six refrigerators, the decision maker must first estimate the importance of the seven standards and rank them, and then, after that, find out each refrigerator’s sorting weight of each standard  separately; finally, combine these information and data to obtain the sorting weight for the total target–the refrigerators to be purchased. With this weight vector, decision making is easy.

Based on the example above, let’s take a look at what the analytic hierarchy process is.

12032副本When a customer purchases a TV set, he considers eight criteria as the evaluation criteria of four TV sets currently on the market and establishes a hierarchical analysis model above.

The highest layer (the purpose of solving the problem); the middle layer (selection of the principles of measures and schemes that must be followed to achieve the overall aim. Also known as the strategy layer, the constraint layer, the rule layer, etc.); The lowest layer (Measures, programs, etc. to solve the problem). Put all the relevant factors to the proper level. Use a hierarchy chart to clearly illustrate the relationship between these factors.

Currently, companies in the industry barely analyze customer consumption so carefully. But for producers, what needs to be considered is actually much more than that. If the needs of consumers can be thought of in advance, and the brand produces items that are easier to sell, and at the same time, the supplier’s responsiveness can keep up, the development of the enterprise is inevitable.

The calculation and implementation of the AHP model require the support of a computer system. At present, not a lot of domestic shoe and clothing enterprises apply it, except some multinational companies. The application of this model can avoid overly personal assessments, performing a comprehensive and multidimensional analysis. It is applicable to industries such as energy system analysis, urban planning, economic management, and scientific research evaluation. But for us, rather than how good this computer system is, the point is to learn this logic through this system to improve our ability to solve problems in practical work and to provide ideas for supplier selection. Besides supplier selection, this idea can be applied as long as it is a multi-variable problem with different priorities; you can even refer to this idea for scoring and selecting, so as to achieve the comprehensive improvement of the industrial chain.




This article is a discussion based on the contents of the following academic journals by the students of Christine Tsui Fashion Buyer Group. The pictures in the article are all from the following academic journal:

Chan, F., & Chan, H. (2010). An AHP model for selection of suppliers in the fast changing fashion market. The International Journal Of Advanced Manufacturing Technology,51(9-12), 1195-1207. doi: 10.1007/s00170-010-2683-6




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s