Chapter 1 Introduction
1.3 Summary of major findings
The major contributions of this study can be summarized as follows. First, we structure the hierarchy of attributes by conducting seven in-depth interviews with industry experts and managers. Second, we survey twenty domain experts to conduct the pairwise comparison to obtain the composite weights. This allows any decision makers in a related industry to use our established model to evaluate her or his shifting supply chain decision.
In the final evaluation stage, we conduct another survey and present our proposed alternatives to three decision makers. We would like to evaluate whether China remains the best place or other alternatives like Taiwan and Vietnam now becomes the better options for operating the supply chain. Our result shows that, at least for now from the viewpoint of the interviewed experts, China is still the most promising place for a foreign company to operate a consumer electronic manufacturing supply chain.
Chapter 2
Literature Review
2.1 Shifting supply chain
Supply chain is the sequence of processes involved in a system of organizations, people, activities, information, and resources to create a product or service from suppliers and deliver to customers. Lehmann (2012) summarizes the shifts beginning with the industrial revolution in Britain, after which the center of global production and trade shifted to Western Europe (especially Germany), then to the U.S., and two decades after World War II to Japan, Asia's Four Tigers (especially South Korea and Taiwan), and then to China.
This observation suggests that shifting supply chain is always considered at any time.
While China has become the world's largest global supply chain leader during the past two to three decades, the US-China trade war in 2008 was severely impacted in some key industries, supply chains are faced to consider moving manufacturing from higher cost to lower cost countries since the companies may foresee more cost-benefit of manufacturing outside of the existing production sites, so-called as shifting supply chain.
Cyrill (2019) mentions the benefits of shifting manufacturing out of China or diversifying production channels may be greater in the longer term given China’s steadily rising labor costs, mounting compliances, social insurance commitments, stringent environmental checks, and other pressures. For example, cable and connector maker Luxshare is setting
up in Vietnam and exploring facilities in India. Wistron, the Taiwanese original design manufacturer, has been calculating the costs of a shift out of China as it factors the US tariff threat to its business. Foxconn, the biggest manufacturer of Apple, announced that it had laid off 50,000 seasonal workers in China since October 2009 due to slowing iPhone sales and is seriously exploring the possibility of manufacturing smartphones to the US and Mexico.
2.2 Analytical hierarchy process
The analytical hierarchy process (AHP) was introduced by Saaty (1986) as a tool to determine the priority of multiple attributes, decomposition quantifiable, and intangible criteria, and evaluate through the pairwise comparison to scale down the selection process.
The applications of AHP ranges from organization planning, performance management to environmental and supplier selection which has been widely used in decision-making aspects.
Saaty (1982) demonstrates a decision-making process for leaders with applying the analytical hierarchy process among 12 pairwise comparison matrices including flexibility, opportunity, security, reputation, salary et cetera. He also constructs a step flow followed in the analytic network decision process to determine the best steps of action in such decision-making process.
Schniederjans and Wilson (1991) present an improved Information System (IS) project selection methodology that combines the analytic hierarchy process (AHP) with a goal programming (GP) model framework. The AHP is first used to prioritize the set of IS projects under consideration on the basis of the pertinent criteria of the organization.
Lai, Trueblood and Wong (1999) discuss how AHP can be applied to the selection of multimedia authoring systems (MAS) in a group decision environment. Six experienced software engineers participated in the study. The results indicate that AHP offers chances for every participant to fully understand, discuss, and objectively evaluate all MAS products before identifying and selecting the most efficient MAS.
Wei, C. C., Chien C.F., & Wang, M.J. (2005) propose a framework to conduct an Enterprise Resource Planning (ERP) system selection based on AHP approach. The framework can systematically construct the objectives of ERP selection to support the business goals and strategies of an enterprise, identify the appropriate attributes, and set up a consistent evaluation standard for facilitating a group decision process.
Chen (2015) evaluates the process of supplier selection in notebook electronic components with a fuzzy AHP method. Based on supplier selection literature, the study analyzes the evaluation dimensions and criteria, invites domain experts and professors to fill in assessment table used the fuzzy Delphi method (FDM) to filter. It then uses the fuzzy analytic hierarchy process (FAHP) to construct the selection model of notebook electronic component supplier. The results show that when managers select suppliers, product quality factor is the most important dimension, and next are cost/production control, innovation, development capabilities and vendor management.
Chen (2019) uses AHP framework to build a performance evaluation process in a memory testing manufacturing industry. It applies an objective hierarchy structure to create the performance indices. Following the AHP to design the investigation and obtain the responses from manufacturing indirect labors to get the weighting between indices.
Furthermore, it uses an empirical case from the memory testing house to verify the validity.
Following the above successful applications of AHP, in this study we also aim to adopt AHP to assess our research objective, shifting supply chain.
Chapter 3
Research Methodology
3.1 Research method
In this study, in order to make a decision in an organized way and to generate priorities, we use the analytical framework of analytical hierarchical process (AHP) to decompose the tangible and intangible key factors and extract the associated attributes to evaluate the multi-criteria of shifting China’s supply chain.
Satty (1986) constructs the AHP model which involves three stages of problem solving: decomposition, comparative judgments, and synthesis of priorities. First, the decomposition principle calls for the construction of a hierarchical network to represent a decision problem, with the top representing overall objectives and the lower levels representing criteria and alternatives.
Next, with the comparative judgments, users are required to set up a comparison matrix at each hierarchy by comparing pairs of criteria. A scale of values ranging from 1 (equal importance) to 9 (extreme importance) are used for expressing user preference.
Last, synthesis of priorities is conducted to calculate a composite weight for each alternative based on preferences derived from the comparison matrix. Once completed
these three stages, it would allow decision makers to obtain a relative standing, on a ratio scale, of the alternatives to be evaluated. The process is illustrated in Figure 3.1.
Figure 3.1: Steps of analysis hierarchy process (modified from Lai et al., 1999)
3.2 Decomposition
In order to structure the critical attributes that involves the shifting supply chain analysis, we conduct seven in-depth interviews with experience consumer electronic manufacturing industry experts and managers in April 2020.
Among the seven industry experts and managers, six have more than ten years work experience in the consumer and business electronic manufacturing industry. All of them have worked on the management of the supply chain in China, as well as managed Taiwan, Vietnam, Thailand, Philippines, South Korea, Japan, US and, Mexico factories in certain timeframe.
Before the official interview, we conduct a pilot survey to test the correctness of the instructions to be measured by whether all the respondents in the pilot sample are able to follow the directions as indicated. From the interviews, we extract each expert’s first-hand knowledge and gather detailed industry knowledge on the determinants of sixteen key attributes. Next, the research team categorize the sixteen attributes into five dimensions including government regulations, society, industry, political stability, nature and present the hierarchy in Figure 3.2. Detailed definitions of all attributes are provided in Table 3.1.
Figure 3.2: Shifting supply chain model Table 3.1: Definition of the attributes Dimensions Attributes Definition
Government Regulations
Business regulation and taxation
The imposition of compulsory levies to the business entities by governments
Industrial incentives and benefits
To what degree does the government provide qualified businesses different kinds of access to some of the available industrial incentive programs
Environmental regulation
To what degree does the government impose environmental regulation that attempts to protect public health and the environment from pollution by industry and development
Tariff The amount of tax or duty to be paid on a
particular class of imports or exports Intellectual
property
To what degree does the government protect intellectual property, which is a category of property that includes intangible creations of the human intellect including copyrights, patents, trademarks, and trade secrets
Society Public
infrastructure
The completeness of infrastructure owned or available for use by the public, including aviation, rail and road transport, energy, water, and telecommunication infrastructure
Labor market The size of labor market, level of average wage, education level, relative industrial skills, and language proficiency
Culture To what degree does the culture (including lifestyle, religion, etc.) impacts employees’ work diligence
Consumer market The maturity of a consumer electronics market Safety How safe it is to run a factory in the given area,
evaluated by the crime rate, absence of violence and terrorism, numbers of refugees, and heavy weapons accessness
Industry Completeness of supply chain
Geographic concentration of a complete supply chain facilitating upstream and downstream flows of products, services, finances and/or information from a source to a customer Thickness of
supply chain
Easiness to find second/third sources in the same tier (measured by the number of suppliers in that tier), providing manufacturers more benefits regarding cost reduction, quality comparison, and risk reduction
Political Stability
Opposite political parties affiliation
The frequency of opposite political decisions made within a country that might result in an unanticipated loss to investors and foreign business
Completeness of legal system
The condition of having all the necessary and appropriate legal system structure (including civil law, common law, and religious law) within a country
Nature Distance of supply chain entities
The overall distance caused by the geographic locations of the entities in the supply chain,
which impacts the material and product delivery logistics lead time
Richness of natural resource
The richness of rare resources like metals, minerals, or others that can be acquired from the ground for electronic components manufacturing
3.3 Comparative judgement
In the following stage, we construct a set of pairwise comparison matrices and conduct twenty surveys from the domain experts to get the result of weighing the priorities within the attributes in each level in May 2020.
To make comparisons, we need a scale of numbers that indicates how many times more important or dominant one element is over another element with respect to the criterion or property for which they are compared. Table 3.2 exhibits the scale.
Table 3.2: The fundamental scale of absolute numbers (modified from Satty, 1986) Intensity of
Importance Definition Explanation
1 Equal Importance Two activities contribute equally to the objective
3 Moderate importance of one over another
Experience and judgement slightly favor one activity over another 5 Essential or strong importance Experience and judgement strongly
favor one activity over another 7 Demonstrated importance An activity is favored very strongly
over another; its dominance demonstrated in practice
9 Extreme importance The evidence favoring one activity over another is of the highest possible order of affirmation
2, 4, 6, 8 Intermediate values between the two adjacent judgements
When compromise is needed
After we get the reported relative importance of each pairwise attributes, we calculate the relative weights by using the eigenvalue method and obtain the consistency index (CI) and consistency ratio (CR) for each paired comparison matrix. The consistency ratio should be less than the threshold value 0.1 to ensure that the decision maker is consistent in assigning paired comparisons. For those CR which are out of the baseline, we follow Satty (1990) to attempt to fix the matrix and recalculate CI and CR. For those CR that still cannot be fixed, we remove those records and use the average of the remaining to generate the relative weights of each attribute.
To obtain the relative weights and overall ranking among all second-level attributes, we multiply the second-level relative weights with its parent first-level weight according to the hierarchy to obtain the composite weights. These composite weights represent the decision makers’ opinion of the relative importance of each decision alternative. The overall ranking may then be derived by comparing the composite weights, where an attribute with a larger composite weight is ranked higher.
3.4 Final evaluation
In the final evaluation stage, we do another survey and present our proposed alternatives to another group of decision makers in June 2020. For each attribute, we ask them to provide their relative weights for each alternative. For each decision maker, we then conduct the same method of the consistency check, remove those whose CR are higher than the threshold value 0.1, and calculate her/his score for each alternative through weighted average. These scores are then averaged among all decision makers to result in the final scores for each alternative.
Chapter 4
Analysis and Results
4.1 Synthesis of priorities
In this chapter, we follow the research method of analytical hierarchy process (AHP) to construct a set of pairwise comparison matrices and conduct twenty surveys from the dominant experts to get the result of weighing the priorities within the attributes in each level. The survey is divided into three parts: basic information, first level of attributes, and second level of attributes.
4.1.1 Overall summary
presents on the demographics of the respondents. We conduct the survey to the industrial experts who have work experience in China and cross-countries management with at least three-year work experience. Among the total twenty responses, out of nineteen are from the consumer electronics manufacturing industry, and one is from the government department in Taiwan. Out of nine respondents have less than ten years work experience and up to eleven have more than ten years work experience. In addition, majority of the
work location of respondents are from Taiwan, and out of six are from China and two are from Southeast Asia who locate in Singapore and Malaysia.
Table 4.1: Demographics of the respondents
Category Items Numbers Percentage
Gender Male 11 55%
Female 9 45%
Age 25-29 6 30%
30-39 8 40%
40-49 6 30%
Education Undergraduate 8 40%
Postgraduate 12 60%
Position Program management 15 75%
Supply chain management 3 15%
Others 2 10%
Work Location Taiwan 12 60%
China 6 30%
Southeast Asia 2 10%
In the second stage, we analyze the pairwise priority comparison of the first-level attributes which include the “Government regulations”, “Society”, “Industry”, “Political stability”, and “Nature”. By calculating the normalized eigenvector and conducting the consistency check, the summary of relative weights for each attribute is shown in Table 4.2. The majority of the respondents consider “Government regulations” (37.6%) is the most important dimension for foreign company to evaluate whether a given area is suitable to run the supply chain or not.
Table 4.2: The pairwise comparison result of the first-level attributes
Attributes Weights Ranking
Government 0.376 1
Industry 0.221 2
Political Stability 0.199 3
Society 0.127 4
Nature 0.076 5
In the third stage, we analyze the pairwise priority comparison of the second-level attributes. First of all, the survey conducts the attributes within “Government regulations”.
Through the same process, the summary of relative weights for each attribute is shown in Table 4.3. The result shows the majority of the respondents consider “Business regulations and taxation” (31.4%) is the most important dimension within government regulations.
Table 4.3: The pairwise comparison results of “Government regulations”
Attributes Weights Ranking
Business regulations and taxation 0.314 1
Tariff 0.261 2
Industrial incentives and benefit 0.245 3
Environmental regulation 0.091 4
Intellectual property 0.089 5
Secondly, the survey conducts the attributes within “Society”. The summary of relative weights for each attribute is shown in Table 4.4. Among all the factors, the majority of the respondents consider “Labor market” (33.8%) is the most important dimension within society. Labor market not only refers to the size of labor market, but also contains the level of average wage, education, relative industrial skills, and language proficiency.
Table 4.4: The pairwise comparison results of “Society”
Attributes Weights Ranking
Labor market 0.338 1
Public infrastructure 0.203 2
Safety 0.201 3
Consumer market 0.168 4
Culture 0.090 5
Thirdly, the survey conducts the attributes within “Industry”. The summary of relative weights for each attribute is shown in Table 4.5. Between the two factors of completeness and thickness of supply chain, the large portions of the respondents consider
“Completeness of supply chain” (72.1%) is a strongly important dimension within industry from the integration of upstream and downstream.
Table 4.5: The pairwise comparison results of “Industry”
Attributes Weights Ranking
Completeness of supply chain 0.721 1
Thickness of supply chain 0.279 2
Next, the survey conducts the attributes within “Political stability”. The summary of relative weights for each attribute is shown in Table 4.6. Between the two factors, the large portions of the respondents consider “Completeness of legal system” (60.3%) is strongly important than the opposite political parties affiliation.
Table 4.6: The pairwise comparison results of “Political stability”
Attributes Weights Ranking
Completeness of legal system 0.603 1
Opposite political parties affiliation 0.397 2
Lastly, the survey conducts the attributes within “Nature”. The summary of relative weights for each attribute is shown in Table 4.7. Between the two factors, the large portions of the respondents consider “Distance of supply chain entities” (71.3%) is strongly important than the richness of natural resource.
Table 4.7: The pairwise comparison results of “Nature”
Attributes Weights Ranking
Distance of supply chain entities 0.713 1
Richness of natural resource 0.287 2
The overall priorities of the first level and second level attributes are shown in Table 4.8. “Completeness of supply chain” (16%) which hits the largest portion of total priority, is considered as the most important dimension among the sixteen factors, following by the “completeness of legal system” (12%) and “Business regulations and taxation” (11.8%).
Table 4.8: Synthesis of the pairwise comparison results (composite) Dimensions Weights
(Rank) Attributes Weights
(Rank)
Composite weights 2
(Rank)
Government 0.376 (1)
Business regulations and
taxation 0.314 (1) 0.118 (3)
Tariff 0.261 (2) 0.098 (4)
Industrial incentives and benefit 0.245 (3) 0.092 (5) Environmental regulation 0.091 (4) 0.034 (10) Intellectual property 0.089 (5) 0.033 (11)
Society 0.127 (4)
Labor market 0.338 (1) 0.043 (9)
Public infrastructure 0.203 (2) 0.026 (12)
Safety 0.201 (3) 0.025 (13)
Consumer market 0.168 (4) 0.021 (15)
Culture 0.09 (5) 0.011 (16)
Industry 0.221 (2)
Completeness of supply chain 0.721 (1) 0.16 (1) Thickness of supply chain 0.279 (2) 0.062 (7) Political
Stability 0.200 (3)
Completeness of legal system 0.603 (1) 0.12 (2) Opposite political parties
affiliation 0.397 (2) 0.079 (6)
Nature 0.076 (5)
Distance of supply chain entities 0.713 (1) 0.054 (8) Richness of natural resource 0.287 (2) 0.022 (14)
Composite weights: Results of multiplying each priority of its attributes and sub-attributes.
4.1.2 Synthesis of priorities (by work experience)
In this study, in order to analyze the difference results among various work experience, we divide the feedback into two major groups. Out of the twenty respondents, the nine having less than ten years of work experience are gathered into the junior group. On the contrary, the other eleven ones are put into the senior group. ranked as the second in the junior and senior groups, respectively.
Table 4.9 and Table 4.10 list the syntheses for these two groups. From these two tables, we may observe that “Completeness of supply chain” hits the largest portion of total priority in both two groups, followed by “Business regulations and taxation” and
“Completeness of legal system” ranked as the second in the junior and senior groups, respectively.
Table 4.9: Synthesis of pairwise comparison within ten years’ work experience
Dimensions Weights
(Rank) Attributes Weights
(Rank)
Composite weights (Rank)
Government 0.426 (1)
Business regulations and taxation 0.33 (6) 0.14 (2)
Tariff 0.253 (8) 0.108 (4)
Industrial incentives and benefit 0.231 (10) 0.098 (5) Environmental regulation 0.105 (14) 0.045 (9) Intellectual property 0.081 (16) 0.034 (10)
Society 0.097 (4)
Labor market 0.342 (5) 0.033 (11)
Public infrastructure 0.235 (9) 0.023 (12)
Consumer market 0.213 (11) 0.021 (13)
Culture 0.107 (13) 0.01 (15)
Safety 0.103 (15) 0.01 (15)
Industry 0.231 (2) Completeness of supply chain 0.716 (2) 0.165 (1) Thickness of supply chain 0.284 (7) 0.066 (7)
Political
Stability 0.188 (3)
Completeness of legal system 0.58 (3) 0.109 (3) Opposite political parties
Completeness of legal system 0.58 (3) 0.109 (3) Opposite political parties