**2. Concept and theory**

reduce and distribute risky. The subcontractor can be classified into several levels namely 2nd tier, 3rd tier, and so on which relied on the role and performance evaluation. To become a subcontractor, a company is applied and audited by a SCM committee of the 1st tier company. The applicants are evaluated by company profile, factory visit, auditing 3 divisions; planning, quality assurance, and supply chain, based on the subcontracting criteria. There are 4 main criteria such as safety, planning, production control, and quality. The safety aims to check that the subcontractor works safely in the well-prepared environment including machine protection, operator protection, and part protection. Planning involves with production supporting process. They are production planning, raw material requirement planning, order receiving plan, compound and components receiving plan, sending raw material back to the 1st tier, delivery plan, packing standard plan, and inventory control. In terms of supply chain, It can be classified into 3 main group based on manufacturing types: sheet metal parts, plastic parts, and rubber parts. However, production technology is not concern in this chapter but it focuses on performance management and evaluation of delivery, back order clearance, and quality of finished parts. One of the most important aspects is that customer satisfaction on the customer service process that suppliers must provide any automotive parts needed whenever they want. In this paper, it is concerned on the case study and information based on the rubber part manufacturing industry. Production process control consists of compound storage control, curing and de-flashing, machine and mold preventive maintenance, and production control. Quality involves with inspection, finished goods and defect management as well as problem solving, quality assur-

*Concepts, Applications and Emerging Opportunities in Industrial Engineering*

ance based on ISO (record, traceability, and change management).

audit includes working system, ISO, improvement, control, and safety

performance and allocate production planning.

environment.

**208**

The evaluation for renew-contract consists of two parts; monthly audit and yearly audit. The monthly audit is evaluated by the performance. They are mainly on delivery on-time, quality of finished goods, back order delivery. The result is achieved by grade level; A, B, C, and D. Grade A is excellent. B is good. C and D are needed to be improved. If the improvement is not successful and unsatisfaction, then the contract is resigned. The more details are in the later section. The yearly

Supply chain performance evaluation for automotive supply chain is quite seriously emphasized on product quality and delivery on-time which are impact on human's live. The general problem is that the market is fluctuate and heavy competitiveness on the cost leadership. Supply chain and subcontractors are the key success methodology which are widely used all over the world. However, the problems are; 1. how to select subcontractors to produce parts effectively; 2. how to maximize cost when the capacities are suitably shared and distributes; 3. how to maximize cost for transportation of the whole subcontractor locations; 4. how to manage machine availability and uncertainty; 5. how to deal with uncertain

Modern computer information technology is currently applied to supply chain

resources planning (ERP), internet of things (IOT), could manufacturing and so on. However, the SME subcontractors are not need to implement the fully high information technology or fully automation but it should invest or develop the parts of technology that can link to accompany the 1st tier company. One of the most critical part of the computer information technology used in the supply chain is could computing and monitoring as well as production execution in order to distribute and share information rapidly, real-time and on-time needed. Delivery is the most important for the automotive subcontractors especially at the 1st tier company. This chapter proposes automotive industrial supply chain performance evaluation under

management such as computer integrated manufacturing (CIM), enterprise

This section presents the concept and theory of supply chain management, fuzzy logic, fuzzy AHP, Neural network and performance evaluation. Supply chain is major links to every parts of the business processes and communicate to their chains and the 1st tier in both vertical and horizontal organization. Fuzzy logic is a methodology used to deal with uncertainty including fuzzy AHP which helps to arrange the priority factors for multi-decision making to reduce complexity. Neural network is used to deal with qualitative data of performance evaluation.

#### **2.1 Supply chain management**

Supply chain plays significant role to OEMs in terms of increasing market share in a highly competition industrial cluster such as automotive industry by accessing to advanced manufacturing technology, reduced time to market, lower production costs, and more effectively used of assets. Modern supply chain management has seamless relationship in dealing with uncertain and fluctuate demands, risk management, stock management, allocate capacity, real-time tracking manufacturing progress with on-line and could system as well as using and selection outsource as engineering service providers. In addition, the customer relationship management is critical for the effective supply chain management. In the previous time, the average time required for company to process and delivery to market and customer from warehouse inventory taken many days or even unpredictable time because of uncertainty. It might be inventory out of stock, misplaced work order, misdirected shipment, total time to service customer escalated rapidly.

Presently, the world changing a lot influencing from digital age and disruption. Therefore, the information technology together with computerization are performed as a backbone of most of the business process according to 4G and 5G even the 6G is coming soon. Orders from customer are rapidly changed to ecommerce and on-line purchasing which is supported by modern logistics controlling by GPS and cloud monitoring. The reality of connectivity among collaborating business organization continues to drive a new order of relationship called supply chain management.

Supply chain is traditionally combined with logistic which consists of two major elements; inbound and out bound logistics. The process begins with second tier suppliers to deliver parts or products to the first-tier suppliers' through manufacturing processes. The inbound logistics are applied in this stage. Then, the finished goods are sent to distributors, agencies, and retails or even end users and customers. The structure of the supply chain is varied based on the companies. Therefore, the unique management is properly designed and created. However, the principle characteristics of supply chain are defined as ability of a firm to work with its suppliers to provide high quality material and components which are competitively priced. The closeness of relationship between vendor and customer in respects purchasing materials and inventory management reflects to the company's strategy and the role of supplier contributing to the long-term success of the firm.

Supply chain management is a well-known concept to applied in the modern business management all over the world particularly in the era of digital disruption and transformation. In the past, the supply chain management is applied to manage material flows among work stations inside the companies accompany with kanban, push-pull system, WIP and buffer inventory control in the JIT system including waste management of the lean manufacturing. Presently, the supply chain management is applied to a group of distributed manufacturing and outbound logistics in and out of the first company. It is recognized to be a critical tool to make survival and competitiveness even up to the competitive advantage of the whole chain. The key success factors are communication, delivery, quality, cost under uncertain circumstance. The digital transportation used is applied the progress of manufacturing processes by on-line and real- time tracking using cloud computing system. The evaluation can be measured by the key success factors in order to achieve the customer satisfaction. On the other hand, the supply chain cluster has to implement lean production in order to dealing with competitiveness problem. Company often had an antagonistic relationship with their suppliers inside the supply chain cluster. Every item that was purchased had several vendors who are played off against each other in order to obtain the lowest possible prices, which was the first criteria for being awarded a contract. Subcontractors recognizing that the relationship could very likely be terminated with the next contract, invested minimal time and money to address the specific needs of individual customers. The purchasing function within the manufacturing company often reported to the supply chain manager in order to make decision on purchasing raw material and components at the lowest possible cost.

decrease of standard time, reduction of equipment downtime, warehouse and transportation labor productivity. Asset is measured by inventory turns, inventory levels, number of days of supply, obsolete inventory, return on net assets, return on

*Automotive Industrial Supply Chain Performance Evaluation under Uncertain Constraints…*

Performance evaluation for supply chain is different from a common firm particularly for subcontractors. There are multi-criteria evaluation in several hierarchy and different weights depending of the importance priorities. The main criteria are delivery on the purchasing orders and on-time and on quality. If any uncertainty is occurred on the delivery, back order must be released on the certain time. The quality involves with defects and errors occurring on the delivery. It is normally measured by six sigma (ppm). Presently, modeling and simulation are used as a manufacturing system tool for production flow and performance evaluation that most of the mathematical models are used in operations management and industrial

**Figure 1** shows the simulation and modeling which applies fuzzy logic and Neuro fuzzy to evaluate the performance of subcontractors in automotive supply chain. It is started on the real-world system. The objectives, constraints, and alternatives are identified. Customer requirements are taken into account for designing and creating measurement parameters and controls for modeling using mathematic model or intelligent system such as fuzzy logic, neural networks and so on. Data at the current and activated factory of the subcontractors while doing real business is

This section explains the previous works on the methodology dealing with performance evaluation in the uncertainty circumstances because of changing world caused by modern communication and management of digital technology. Fuzzy

collected. Then the performance evaluation methodology is designed and implemented. On the other hand, the simulation and modeling system are developed. Fuzzy logic and neuro fuzzy system are formulated with membership functions, fuzzification and defuzzification. Finally, the performance evaluation

modeling system are created and implemented.

investment, economical value added (EVA) and so on.

*Simulation and modeling framework for performance evaluation.*

*DOI: http://dx.doi.org/10.5772/intechopen.93679*

engineering.

**Figure 1.**

**3. Related works**

**211**

### **2.2 Performance evaluation**

Performance of production is the final goal of manufacturing which needs to measure for evaluation at the end of processes or periods of time. It can consider on quantity, quality, cost, value, productivity, resource used and so on. Effective performance evaluation for competitive advantage particularly in supply chain logistics includes monitoring, controlling, and directing operations. Monitoring is accomplished by the establishment of appropriate metrics to track system performance for reporting to management such as on-time delivery, production rate, production quality on planned, finished goods, defects, transportation tracking, warehousing and so on [1]. Controlling is accomplished by having appropriate standards of the whole life cycle of operation performance related to the customer requirements or international standard (ISO). Controlling is also used to ensure that planning time is managed effectively on the productive operations. Directing is related to shop floor control on the machine level and operators to run the production properly, correctly, and perfectly in order to achieve high level of productivity.

Performance metrics is typically involved with several criteria; cost, customer services, quality, productivity and asset. Costs are cost per unit, total cost, percentage cost of sales, administrative coat, direct labor cost, inventory carrying cost, cost of return goods, cost of defect, cost of damage, cost of service failures, cost of back order, cost of logistics, cost of materials and so on. Criteria of customer services are on-time delivery, back orders, cycle time, complete orders, delivery consistency, fill rate stockouts, response accuracy, customer complaints, reliability, overall satisfaction. Quality is the most criteria consisting of product quality, order entry accuracy, invoicing accuracy, information accuracy, number of customer complaints, number of customer returns, picking and shipping accuracy. Productivity criteria are decrease cost rate, number of increase production quantity rate, productivity index, order per sales representation, units per labor values, units shipped per employee,

*Automotive Industrial Supply Chain Performance Evaluation under Uncertain Constraints… DOI: http://dx.doi.org/10.5772/intechopen.93679*

#### **Figure 1.**

Supply chain management is a well-known concept to applied in the modern business management all over the world particularly in the era of digital disruption and transformation. In the past, the supply chain management is applied to manage material flows among work stations inside the companies accompany with kanban, push-pull system, WIP and buffer inventory control in the JIT system including waste management of the lean manufacturing. Presently, the supply chain management is applied to a group of distributed manufacturing and outbound logistics in and out of the first company. It is recognized to be a critical tool to make survival and competitiveness even up to the competitive advantage of the whole chain. The key success factors are communication, delivery, quality, cost under uncertain circumstance. The digital transportation used is applied the progress of

*Concepts, Applications and Emerging Opportunities in Industrial Engineering*

manufacturing processes by on-line and real- time tracking using cloud computing system. The evaluation can be measured by the key success factors in order to achieve the customer satisfaction. On the other hand, the supply chain cluster has to implement lean production in order to dealing with competitiveness problem. Company often had an antagonistic relationship with their suppliers inside the supply chain cluster. Every item that was purchased had several vendors who are played off against each other in order to obtain the lowest possible prices, which was the first criteria for being awarded a contract. Subcontractors recognizing that the relationship could very likely be terminated with the next contract, invested minimal time and money to address the specific needs of individual customers. The purchasing function within the manufacturing company often reported to the supply chain manager in order to make decision on purchasing raw material and

Performance of production is the final goal of manufacturing which needs to measure for evaluation at the end of processes or periods of time. It can consider on quantity, quality, cost, value, productivity, resource used and so on. Effective performance evaluation for competitive advantage particularly in supply chain logistics includes monitoring, controlling, and directing operations. Monitoring is accomplished by the establishment of appropriate metrics to track system performance for reporting to management such as on-time delivery, production rate, production quality on planned, finished goods, defects, transportation tracking, warehousing and so on [1]. Controlling is accomplished by having appropriate standards of the whole life cycle of operation performance related to the customer requirements or international standard (ISO). Controlling is also used to ensure that planning time is managed effectively on the productive operations. Directing is related to shop floor control on the machine level and operators to run the production properly, correctly, and perfectly in order to achieve high level of productivity. Performance metrics is typically involved with several criteria; cost, customer services, quality, productivity and asset. Costs are cost per unit, total cost, percentage cost of sales, administrative coat, direct labor cost, inventory carrying cost, cost of return goods, cost of defect, cost of damage, cost of service failures, cost of back order, cost of logistics, cost of materials and so on. Criteria of customer services are on-time delivery, back orders, cycle time, complete orders, delivery consistency, fill rate stockouts, response accuracy, customer complaints, reliability, overall satisfaction. Quality is the most criteria consisting of product quality, order entry accuracy, invoicing accuracy, information accuracy, number of customer complaints, number of customer returns, picking and shipping accuracy. Productivity criteria are decrease cost rate, number of increase production quantity rate, productivity index, order per sales representation, units per labor values, units shipped per employee,

components at the lowest possible cost.

**2.2 Performance evaluation**

**210**

*Simulation and modeling framework for performance evaluation.*

decrease of standard time, reduction of equipment downtime, warehouse and transportation labor productivity. Asset is measured by inventory turns, inventory levels, number of days of supply, obsolete inventory, return on net assets, return on investment, economical value added (EVA) and so on.

Performance evaluation for supply chain is different from a common firm particularly for subcontractors. There are multi-criteria evaluation in several hierarchy and different weights depending of the importance priorities. The main criteria are delivery on the purchasing orders and on-time and on quality. If any uncertainty is occurred on the delivery, back order must be released on the certain time. The quality involves with defects and errors occurring on the delivery. It is normally measured by six sigma (ppm). Presently, modeling and simulation are used as a manufacturing system tool for production flow and performance evaluation that most of the mathematical models are used in operations management and industrial engineering.

**Figure 1** shows the simulation and modeling which applies fuzzy logic and Neuro fuzzy to evaluate the performance of subcontractors in automotive supply chain. It is started on the real-world system. The objectives, constraints, and alternatives are identified. Customer requirements are taken into account for designing and creating measurement parameters and controls for modeling using mathematic model or intelligent system such as fuzzy logic, neural networks and so on. Data at the current and activated factory of the subcontractors while doing real business is collected. Then the performance evaluation methodology is designed and implemented. On the other hand, the simulation and modeling system are developed. Fuzzy logic and neuro fuzzy system are formulated with membership functions, fuzzification and defuzzification. Finally, the performance evaluation modeling system are created and implemented.

### **3. Related works**

This section explains the previous works on the methodology dealing with performance evaluation in the uncertainty circumstances because of changing world caused by modern communication and management of digital technology. Fuzzy

logic is firstly explained following with neuro-fuzzy system. Then previous works on performance evaluation particularly on supply chain management is reviewed and comments. Finally, the modern management on cloud monitoring is stated and discussed to be the way of applying to the monitor performance operation at shop floor level.

#### **3.1 Fuzzy logic**

Fuzzy logic is very useful tool to support decision making under uncertainty and complexity when information is not sufficient and imprecision. Traditionally, probability theory is used to deal with uncertainty. However, the uncertainty can be different forms. The probability can deal with the expectation of the future event based on something known before. By contrast, fuzzy logic handles with a prediction about the event which represents fuzziness expression in terms of linguistic [2]. Zadeh [3] first presented a Fuzzy Set theory. It is a tool that helps in decision making under the uncertainty of data. The fuzzy logic has flexibility and can make easy decision making by applied in the representation of human reasoning and linguistic terms with the following definitions:

#### *3.1.1 Definition 1. Fuzzy set*

A fuzzy set is a member function that attributes the elements of the domain in the interval [0,1] as the following Eq. (1). The interval is a degree of membership when the given value is close to 1, there will be a higher level of membership. If the membership is zero, it means that there is no membership.

$$\mathbf{U\_{a}(X) : U \to [0, 1]} \tag{1}$$

*3.1.5 Definition 5. Defuzzification*

*DOI: http://dx.doi.org/10.5772/intechopen.93679*

*Triangular membership function.*

**Figure 2.**

can be done and substitute the fuzzy logic.

**213**

**3.2 Artificial neural network and neuro fuzzy system**

ANN mimics human's brain working using algorithms and graph theory. It represents biological nervous systems such as brain processing information deriving

A defuzzification operation is a process to transform fuzzy set into a crisp output. The center of gravity (COG) is the most commonly technique calculating the center of the area of combined membership function. In this paper, fuzzy logic

*Automotive Industrial Supply Chain Performance Evaluation under Uncertain Constraints…*

<sup>P</sup> *<sup>y</sup>* <sup>¼</sup> *<sup>y</sup>*<sup>0</sup> j j � *<sup>y</sup>*<sup>0</sup>

Previous works have been done using fuzzy logic for operation performance evaluation particularly in the area of supply chain management. Fuzzy logic for logistics management using SCOR model for evaluating performance in supply chain is widely studied [4–6]. The model developed can evaluate overall performance by real-time network. The model can predict the performance based on causal relationship of the SCOR metrics using fuzzy rules to build the prediction model. The fuzzy logic model can be combined with Linear Programming (LP) which aimed to evaluate the performance and capability in order to distribute purchasing orders in the cluster environment. The study presented the cluster body manufacturing in Thailand [7]. The cluster consists of 40 SME companies which located in the same areas. The have limitation of manufacturing resources. Therefore, they have to combine capacity in order to serve the big lots of purchasing orders. It is normally order with big lots of busses particularly the order from local government. For example, the 2016 purchasing orders were around 500 city busses with limited delivery time. The cluster of bus body manufacturing needs to join and share capacity, resources and even profit. The model develop was tested and help lots of benefits to them. The KPI indicators of the efficiency performance evaluation which is designed with 5 criteria; capacity, quality, reliability, flexibility and source. The fuzzy logic is extended to integrated with other methods such as Quality Function Deployment (QFD) to use for quality design of large complex products and structures. Although the fuzzy logic is effective but it has got a weakness in term of prediction by learning. In this case, the Artificial Neural Network (ANN)

<sup>P</sup> *<sup>y</sup>* <sup>¼</sup> *<sup>y</sup>*<sup>0</sup> j j (2)

sugeno type was used to approximate output as the following Eq. (2):

*y* ¼

In which *U*<sup>a</sup> (x): U ! [0,1] is called pertinence function and *U*<sup>a</sup> (x) is the degree of pertinence of x.

### *3.1.2 Definition 2. Linguistic variable*

A Fuzzy set can be used to describe the value of a variable. A linguistic variable is a linguistic expression which is used to determine the value of what is described in terms of qualitative scales such as very low, low, medium, high and very high.

### *3.1.3 Definition 3. Fuzzification*

The fuzzification is used to convert the input into fuzzy variables or fuzzy sets or language variables.

#### *3.1.4 Definition 4. Membership function*

The membership function is the process of determining the membership level of a variable, which is important for the process of thinking and solving problems. The membership functions are not symmetric or symmetrical in all respects. The membership function used in this paper is triangular. The triangular membership function has a triangle shape, which depends on the 3 variable values, a, b and c as shown in **Figure 2**. It is a commonly used in many researches.

*Automotive Industrial Supply Chain Performance Evaluation under Uncertain Constraints… DOI: http://dx.doi.org/10.5772/intechopen.93679*

**Figure 2.** *Triangular membership function.*

logic is firstly explained following with neuro-fuzzy system. Then previous works on performance evaluation particularly on supply chain management is reviewed and comments. Finally, the modern management on cloud monitoring is stated and discussed to be the way of applying to the monitor performance operation at shop

*Concepts, Applications and Emerging Opportunities in Industrial Engineering*

Fuzzy logic is very useful tool to support decision making under uncertainty and

A fuzzy set is a member function that attributes the elements of the domain in the interval [0,1] as the following Eq. (1). The interval is a degree of membership when the given value is close to 1, there will be a higher level of membership. If the

In which *U*<sup>a</sup> (x): U ! [0,1] is called pertinence function and *U*<sup>a</sup> (x) is the degree

A Fuzzy set can be used to describe the value of a variable. A linguistic variable is a linguistic expression which is used to determine the value of what is described in terms of qualitative scales such as very low, low, medium, high and very high.

The fuzzification is used to convert the input into fuzzy variables or fuzzy sets or

The membership function is the process of determining the membership level of a variable, which is important for the process of thinking and solving problems. The membership functions are not symmetric or symmetrical in all respects. The membership function used in this paper is triangular. The triangular membership function has a triangle shape, which depends on the 3 variable values, a, b and c as

shown in **Figure 2**. It is a commonly used in many researches.

Uað Þ X : U ! ½ � 0, 1 (1)

complexity when information is not sufficient and imprecision. Traditionally, probability theory is used to deal with uncertainty. However, the uncertainty can be different forms. The probability can deal with the expectation of the future event based on something known before. By contrast, fuzzy logic handles with a prediction about the event which represents fuzziness expression in terms of linguistic [2]. Zadeh [3] first presented a Fuzzy Set theory. It is a tool that helps in decision making under the uncertainty of data. The fuzzy logic has flexibility and can make easy decision making by applied in the representation of human reasoning and

linguistic terms with the following definitions:

membership is zero, it means that there is no membership.

*3.1.1 Definition 1. Fuzzy set*

of pertinence of x.

language variables.

**212**

*3.1.2 Definition 2. Linguistic variable*

*3.1.3 Definition 3. Fuzzification*

*3.1.4 Definition 4. Membership function*

floor level.

**3.1 Fuzzy logic**

#### *3.1.5 Definition 5. Defuzzification*

A defuzzification operation is a process to transform fuzzy set into a crisp output. The center of gravity (COG) is the most commonly technique calculating the center of the area of combined membership function. In this paper, fuzzy logic sugeno type was used to approximate output as the following Eq. (2):

$$\mathbf{y} = \frac{\sum |\mathbf{y} = \mathbf{y}'| \times \mathbf{y}'}{\sum |\mathbf{y} = \mathbf{y}'|} \tag{2}$$

Previous works have been done using fuzzy logic for operation performance evaluation particularly in the area of supply chain management. Fuzzy logic for logistics management using SCOR model for evaluating performance in supply chain is widely studied [4–6]. The model developed can evaluate overall performance by real-time network. The model can predict the performance based on causal relationship of the SCOR metrics using fuzzy rules to build the prediction model. The fuzzy logic model can be combined with Linear Programming (LP) which aimed to evaluate the performance and capability in order to distribute purchasing orders in the cluster environment. The study presented the cluster body manufacturing in Thailand [7]. The cluster consists of 40 SME companies which located in the same areas. The have limitation of manufacturing resources. Therefore, they have to combine capacity in order to serve the big lots of purchasing orders. It is normally order with big lots of busses particularly the order from local government. For example, the 2016 purchasing orders were around 500 city busses with limited delivery time. The cluster of bus body manufacturing needs to join and share capacity, resources and even profit. The model develop was tested and help lots of benefits to them. The KPI indicators of the efficiency performance evaluation which is designed with 5 criteria; capacity, quality, reliability, flexibility and source. The fuzzy logic is extended to integrated with other methods such as Quality Function Deployment (QFD) to use for quality design of large complex products and structures. Although the fuzzy logic is effective but it has got a weakness in term of prediction by learning. In this case, the Artificial Neural Network (ANN) can be done and substitute the fuzzy logic.

#### **3.2 Artificial neural network and neuro fuzzy system**

ANN mimics human's brain working using algorithms and graph theory. It represents biological nervous systems such as brain processing information deriving from imprecise and complicated data. Training ANN can be used to provide projections given new situations [8]. Kocamaz et al. [9] proposed ANN which was used to control chaotic supply chain based on a nonlinear dynamic system. It is sensitive dependence on initial conditions and involves with infinite number of different periodic responses. Zayegh and Bassam [10] presented neural network principals and applications. ANN can be used to implement different stages of processing systems based on leaning algorithms by controlling their weights and biases. The paper presented the ANN application for digital signal processing performed by the concept of a multilayer perceptron which feeding forward as networking system by a set of neurons together with weights. It consists of an input layer, multi hidden layers and output layer. Back propagation is an algorithm working together with the multilayer perceptron as shown in the **Figure 3**.

Thipparat [18] proposed application of adaptive neuro fuzzy inference system in supply chain management evaluation in the case study of construction project. The project consists of design, contract, liabilities, weather, soil conditions, environment, and so on which are uncertain. The construction project often deals with many parties of stakeholders in supply chain. The criteria are taken into account

*Automotive Industrial Supply Chain Performance Evaluation under Uncertain Constraints…*

Evaluation is needed to all subcontractors every month and every year in order to check capability, performance and availability to continuous the business. The major criteria are delivery finished goods and all item according to the purchasing order. The 1st tier company plays responsibility to the OEM which is to delivery any part their needs on demands uncertainly. The facing troubles are that demands are fluctuated based on the world economic situations. Therefore, the real-time monitoring is needed in order to dealing with rapid change and effective control stock of inventory. Mahmood [19], Kaganski [20], Amrina & Yusof [21] proposed the concept of production evaluation in SME network using virtual enterprise. Key performance indicators (KPIs) is created which affect operation such as efficiency, utilization, and productivity. Hon [22] proposed the KPIs of production performance consisting of 5 criteria; quality, cost, delivery, and flexibility. Performance evaluation with KPIs is a good tool for controlling and enhancing the company to improve productivity and competitiveness but the methodology cannot use the same criteria particularly for evaluating a supply chain performance. Behrouzi et al. [23] presented the performance evaluation for supply chain management in the automotive industry. It consists of 20 KPIs which applied lean concepts in the context. The studied collected data from the 133 supply chain companies. The criteria contain attributes; waste elimination, continuous improvement, just in time, and flexibility. Roda & Macchi [24] proposed an evaluation model for production system based on the need of factory level performance metric tracking system which adopted OEE criteria to create KPIs. Joppena et al. [25] introduced a KPIs for production performance evaluation consisting of 5 criteria; quality rate, manufacturing defect rate, rework rate, rejection rate and OEE. Hyytia and Rgihter [26] proposed a performance evaluation using simulation modeling for dispatching systems of the routing jobs to the work stations using parallel computing systems. The system can deal with dynamic dispatching policy accompanied with monte Carlo methodology. In supply chain, the problem sometimes needs to deal with capacity sharing in order to increase potential a large order quantity which is benefited to SME industrial cluster. Butdee and Nitnara [8] proposed a fuzzy logic combined with linear programming (LP) modeling for the cluster capacity and performance evaluation to distribute purchased order suitably for each supply chain cluster of bus body manufacturing in Thailand. The modeling can deal with uncertainty situations when demands are fluctuated. The criteria include capacity, quality, reliability, sources, and flexibility. Jagan et al. [27] reviewed a supply chain performance of the whole production system using the concept of Balance Score Card combined with SCOR model as well as AHP model. Hudson et al. [28] proposed the theory and practice of performance measurement systems for SMEs using multi-criteria such as quality, flexibility, time, finance, customer satisfaction, and human resources. Svalina et al. [29] proposed a neuro fuzzy system for evaluation surface roughness with minimize machining time and maximize material removal rate, recommendation optimal cutting parameters under alternated possibility controlling of the machining processes. The system can deal with complexity, imprecision and uncertainty environment. The developed AHP and Fuzzy AHP

such as cost, asset, flexibility, reliability, responsiveness.

**3.3 Sub-contractor evaluation**

*DOI: http://dx.doi.org/10.5772/intechopen.93679*

**215**

Feng et al. [11] proposed the application of ANN to solve the problem of job shop scheduling using MLP networks. Job shop scheduling involves with several tasks varying from customer orders. A job shop consists of many jobs which are assigned to perform on many machines. ANN is advantage on learning by training approach and mostly deal with qualitative information. However, there are various types of application in manufacturing need to deal with the combination of quantitative and qualitative information. Previous articles stated that neuro-fuzzy can deal with the combination information effectively. There are many applications of ANN to production manufacturing such as Lee and Shaw [12] applied a neural network to deal with a real-time floe shop sequencing. Che [13] proposed the ANN for estimation cost of plastic injection molding. Efendigil and Onut [14] proposed the neurofuzzy based methodology to make multi-decision on multi-stage supply chains which integrates from customers to suppliers through warehouses, retailers, and factory. The ANFIS is used for making decision as artificial neural fuzzy inference system on the customer demands. It consists of several procedures; fuzzification, rule antecedent, normalization, rule consequent, and rule interface. The output carries out with new demand data. Chupin et al. [15] proposed neuro-fuzzy model in supply chain management for object state assessing in the conditions of uncertainty-based supply chain strategy. Didehkhani et al. [16] presented assessing flexibility in supply chain using adaptive neuro fuzzy inference system in order to consider the factors of competitiveness of the world class manufacturing. The criteria include speed, variety, flexibility and integration in production main line. Sremac et al. [17] proposed neuro-fuzzy inference systems approach to decision support system for economic order quantity in supply chain management based on a dynamic situation of information flow, products and funds among different participants. SCM is a complex process and mostly involving uncertainty.

**Figure 3.** *A multilayer perceptron of neural network [11].*

*Automotive Industrial Supply Chain Performance Evaluation under Uncertain Constraints… DOI: http://dx.doi.org/10.5772/intechopen.93679*

Thipparat [18] proposed application of adaptive neuro fuzzy inference system in supply chain management evaluation in the case study of construction project. The project consists of design, contract, liabilities, weather, soil conditions, environment, and so on which are uncertain. The construction project often deals with many parties of stakeholders in supply chain. The criteria are taken into account such as cost, asset, flexibility, reliability, responsiveness.

#### **3.3 Sub-contractor evaluation**

from imprecise and complicated data. Training ANN can be used to provide projections given new situations [8]. Kocamaz et al. [9] proposed ANN which was used to control chaotic supply chain based on a nonlinear dynamic system. It is sensitive dependence on initial conditions and involves with infinite number of different periodic responses. Zayegh and Bassam [10] presented neural network principals and applications. ANN can be used to implement different stages of processing systems based on leaning algorithms by controlling their weights and biases. The paper presented the ANN application for digital signal processing performed by the concept of a multilayer perceptron which feeding forward as networking system by a set of neurons together with weights. It consists of an input layer, multi hidden layers and output layer. Back propagation is an algorithm working together with the

*Concepts, Applications and Emerging Opportunities in Industrial Engineering*

Feng et al. [11] proposed the application of ANN to solve the problem of job shop scheduling using MLP networks. Job shop scheduling involves with several tasks varying from customer orders. A job shop consists of many jobs which are assigned to perform on many machines. ANN is advantage on learning by training approach and mostly deal with qualitative information. However, there are various types of application in manufacturing need to deal with the combination of quantitative and qualitative information. Previous articles stated that neuro-fuzzy can deal with the combination information effectively. There are many applications of ANN to production manufacturing such as Lee and Shaw [12] applied a neural network to deal with a real-time floe shop sequencing. Che [13] proposed the ANN for estimation cost of plastic injection molding. Efendigil and Onut [14] proposed the neurofuzzy based methodology to make multi-decision on multi-stage supply chains which integrates from customers to suppliers through warehouses, retailers, and factory. The ANFIS is used for making decision as artificial neural fuzzy inference system on the customer demands. It consists of several procedures; fuzzification, rule antecedent, normalization, rule consequent, and rule interface. The output carries out with new demand data. Chupin et al. [15] proposed neuro-fuzzy model

in supply chain management for object state assessing in the conditions of

uncertainty-based supply chain strategy. Didehkhani et al. [16] presented assessing flexibility in supply chain using adaptive neuro fuzzy inference system in order to consider the factors of competitiveness of the world class manufacturing. The criteria include speed, variety, flexibility and integration in production main line. Sremac et al. [17] proposed neuro-fuzzy inference systems approach to decision support system for economic order quantity in supply chain management based on a dynamic situation of information flow, products and funds among different participants. SCM is a complex process and mostly involving uncertainty.

multilayer perceptron as shown in the **Figure 3**.

**Figure 3.**

**214**

*A multilayer perceptron of neural network [11].*

Evaluation is needed to all subcontractors every month and every year in order to check capability, performance and availability to continuous the business. The major criteria are delivery finished goods and all item according to the purchasing order. The 1st tier company plays responsibility to the OEM which is to delivery any part their needs on demands uncertainly. The facing troubles are that demands are fluctuated based on the world economic situations. Therefore, the real-time monitoring is needed in order to dealing with rapid change and effective control stock of inventory. Mahmood [19], Kaganski [20], Amrina & Yusof [21] proposed the concept of production evaluation in SME network using virtual enterprise. Key performance indicators (KPIs) is created which affect operation such as efficiency, utilization, and productivity. Hon [22] proposed the KPIs of production performance consisting of 5 criteria; quality, cost, delivery, and flexibility. Performance evaluation with KPIs is a good tool for controlling and enhancing the company to improve productivity and competitiveness but the methodology cannot use the same criteria particularly for evaluating a supply chain performance. Behrouzi et al. [23] presented the performance evaluation for supply chain management in the automotive industry. It consists of 20 KPIs which applied lean concepts in the context. The studied collected data from the 133 supply chain companies. The criteria contain attributes; waste elimination, continuous improvement, just in time, and flexibility. Roda & Macchi [24] proposed an evaluation model for production system based on the need of factory level performance metric tracking system which adopted OEE criteria to create KPIs. Joppena et al. [25] introduced a KPIs for production performance evaluation consisting of 5 criteria; quality rate, manufacturing defect rate, rework rate, rejection rate and OEE. Hyytia and Rgihter [26] proposed a performance evaluation using simulation modeling for dispatching systems of the routing jobs to the work stations using parallel computing systems. The system can deal with dynamic dispatching policy accompanied with monte Carlo methodology. In supply chain, the problem sometimes needs to deal with capacity sharing in order to increase potential a large order quantity which is benefited to SME industrial cluster. Butdee and Nitnara [8] proposed a fuzzy logic combined with linear programming (LP) modeling for the cluster capacity and performance evaluation to distribute purchased order suitably for each supply chain cluster of bus body manufacturing in Thailand. The modeling can deal with uncertainty situations when demands are fluctuated. The criteria include capacity, quality, reliability, sources, and flexibility. Jagan et al. [27] reviewed a supply chain performance of the whole production system using the concept of Balance Score Card combined with SCOR model as well as AHP model. Hudson et al. [28] proposed the theory and practice of performance measurement systems for SMEs using multi-criteria such as quality, flexibility, time, finance, customer satisfaction, and human resources. Svalina et al. [29] proposed a neuro fuzzy system for evaluation surface roughness with minimize machining time and maximize material removal rate, recommendation optimal cutting parameters under alternated possibility controlling of the machining processes. The system can deal with complexity, imprecision and uncertainty environment. The developed AHP and Fuzzy AHP

modeling to deal with constraints and dynamic situations as well as uncertainty risk management.

interfaced tool management, service catalogs, network and storage configurations, and service governors, high performance management [32]. Self-service is the portal that interfaces to log and manage on infrastructure easily for configurations and execution deployments via a templates and customization. Visualization system image is enabled you to access different kind of system images that can be chosen and deployed via the self-service feature. Workload optimization can link to work anywhere seamlessly. It can work automatically or support decision making. It can manage resources better under policy and management rules. For example, the system server is used up to 60% or 75% with 2 GB of the ram. Interface to other management tools is to be a centralization of the data for monitoring overall information and work flow and deployment automation both existing technology and new equipment. This feature can assist the potential of execution system. Service catalogs is allowed users to choose service templates or configurations to apply any service in the catalogs via a list of promotion packages. In addition, users can create and developed their own template in different service categories. Network and storage resource configuration permit users to define several types of storage and networks even set up new storage configuration and sharing bloc level. Service governors are a smart tool which help users understand IT system inside company.

*Automotive Industrial Supply Chain Performance Evaluation under Uncertain Constraints…*

The cloud has to support high performance management for gathering and collecting policies from the services based on infrastructure pattern use. It can configure automatically and deploy services using the right amount of resources

There are a lot of application using cloud system in different domains; automotive industry for process improvements and cost reduction, pharmaceutical production data for down time analysis, medical devices monitoring, education and training, production data for real time monitoring, food and beverage for OEE

Subcontractors are operated under the supply chain management which are closely linked to the 1st tier company. The most important aspect is that the subcontractors must understand the role of collaborative working. The main responsibility of the subcontractors is to receive purchasing order and delivery plan. The **Figure 4** shows the cycle of operations. It starts from receiving parts and delivery order. The raw material requirement planning is done by subcontractors. The production plan is created immediately after finished good stock is checked. Raw material is critical aspect when synthetic rubber is taken into account because its life is limited. It needs to be well-organized and can deal with uncertain environments. The raw material is ordered from the 1st tier company which all volume is combined from several subcontractors when the same material is used. This is the first advantage of the subcontractor supply chain which can share critical scale of resource by obtaining cost reduction. The production can start suddenly when the raw material is arrived. However, the material needs to test beforehand according to the standard specification and operation standard procedure given by the engineering department. The production consists of three operations before delivery to the customer of the 1st tier according to the delivery plan. The production also must be well-prepared for preventive maintenance both machines and molds. The molds are design accompanied with new part model before they are distributed to the subcontractors. This point is the 2nd advantage of the subcontractor supply chain

and give high performance.

analysis to increase productivity.

**4.1 Subcontractor operation**

**217**

**4. Research system methodology**

*DOI: http://dx.doi.org/10.5772/intechopen.93679*

Performance evaluation is currently involved with energy consumption which is enable to save production cost. Energy evaluation is concerned with the energy used for machines, oven, welding machines and so on. The energy consumption management is taken into account together with manufacturing efficiency. Energy is one of the major costs of automotive supply chain particularly the rubber part manufacturing. Most of the compression machines, injection machines are used the heaters to heat rubber materials and continuously heated whole day and whole weeks. It is sometimes worked around 20 days or more per month. Therefore, the energy is critical for the company in supply chine because cost is the major criterion of the competition. Energy control is dynamic and uncertain in aspects of obtaining the OEE of s firm. Scientific methods are applied to deal with the energy control. AHP and Fuzzy AHP including Fuzzy FMEA or even the combination of them. The criteria concern with the cause of uncertainty and risk such as severity, occurrence, and detection. The major goals of production control are cost control, improving quality, and on-time delivery.

#### **3.4 Cloud management system**

Cloud manufacturing system is emerged which is developed from the advanced manufacturing systems using the information and computer technologies such as cloud computing, internet of things (IOT), virtualization and service-oriented technologies, virtualization, cyber physical system. It transforms manufacturing resources and manufacturing capabilities into manufacturing services, which can be managed and operated in an intelligent and unified way to enable sharing facilities. The cloud manufacturing system can provide efficient and reliable, high quality, cost saving control. The system can enhance in the decision-making process while the material flow is tracked following the production process particularly at the final operation of the supply chain which needs to control on-time delivery. Problem solving can be rapidly done even immediately to increase productivity. The collaborative manufacturing management can be done along the whole life cycle of production. Few papers studied and presented could computing, cloud-based infrastructure. Lin Zhang et al. [30] expressed the cloud manufacturing which is the combination of cloud computing, IOT, serviced-oriented technologies and highperformance computing in order to solve the bottlenecks in the informatization development and manufacturing applications. It consists of three core components such as resources, cloud service and manufacturing on cloud. The evolutionary of modern manufacturing started from computer integrated manufacturing system (CIM). Then, the agile manufacturing is established to deal with time to market and supply chain management. It can support the responsiveness quickly while cost and quality can be consistently controlled. Afterward, concurrent engineering and collaborative design are developed to save time to market which was the life cycle of product taken into account. Peng Wang [31] explained the benefits and limitations of could computing for cloud manufacturing. Cloud computing is recognized at the moment that is an aggregate computing resources and a service of things. Presently, services on demand is dramatically increased when the world is in the digital disruption and transformation. Cloud computing is enabled to transition of computation in various forms such as platform, hardware, and software. The benefits of the cloud computing are to obtain lower start-up and operating costs, to ease of access and scalability, to reduce risk on resource provision.

The characteristics of cloud management system are to envision for any feature of IOT such as self-service, visualization of system image, workload optimization,

*Automotive Industrial Supply Chain Performance Evaluation under Uncertain Constraints… DOI: http://dx.doi.org/10.5772/intechopen.93679*

interfaced tool management, service catalogs, network and storage configurations, and service governors, high performance management [32]. Self-service is the portal that interfaces to log and manage on infrastructure easily for configurations and execution deployments via a templates and customization. Visualization system image is enabled you to access different kind of system images that can be chosen and deployed via the self-service feature. Workload optimization can link to work anywhere seamlessly. It can work automatically or support decision making. It can manage resources better under policy and management rules. For example, the system server is used up to 60% or 75% with 2 GB of the ram. Interface to other management tools is to be a centralization of the data for monitoring overall information and work flow and deployment automation both existing technology and new equipment. This feature can assist the potential of execution system. Service catalogs is allowed users to choose service templates or configurations to apply any service in the catalogs via a list of promotion packages. In addition, users can create and developed their own template in different service categories. Network and storage resource configuration permit users to define several types of storage and networks even set up new storage configuration and sharing bloc level. Service governors are a smart tool which help users understand IT system inside company. The cloud has to support high performance management for gathering and collecting policies from the services based on infrastructure pattern use. It can configure automatically and deploy services using the right amount of resources and give high performance.

There are a lot of application using cloud system in different domains; automotive industry for process improvements and cost reduction, pharmaceutical production data for down time analysis, medical devices monitoring, education and training, production data for real time monitoring, food and beverage for OEE analysis to increase productivity.

### **4. Research system methodology**

#### **4.1 Subcontractor operation**

Subcontractors are operated under the supply chain management which are closely linked to the 1st tier company. The most important aspect is that the subcontractors must understand the role of collaborative working. The main responsibility of the subcontractors is to receive purchasing order and delivery plan. The **Figure 4** shows the cycle of operations. It starts from receiving parts and delivery order. The raw material requirement planning is done by subcontractors. The production plan is created immediately after finished good stock is checked. Raw material is critical aspect when synthetic rubber is taken into account because its life is limited. It needs to be well-organized and can deal with uncertain environments. The raw material is ordered from the 1st tier company which all volume is combined from several subcontractors when the same material is used. This is the first advantage of the subcontractor supply chain which can share critical scale of resource by obtaining cost reduction. The production can start suddenly when the raw material is arrived. However, the material needs to test beforehand according to the standard specification and operation standard procedure given by the engineering department. The production consists of three operations before delivery to the customer of the 1st tier according to the delivery plan. The production also must be well-prepared for preventive maintenance both machines and molds. The molds are design accompanied with new part model before they are distributed to the subcontractors. This point is the 2nd advantage of the subcontractor supply chain

modeling to deal with constraints and dynamic situations as well as uncertainty risk

*Concepts, Applications and Emerging Opportunities in Industrial Engineering*

Performance evaluation is currently involved with energy consumption which is enable to save production cost. Energy evaluation is concerned with the energy used for machines, oven, welding machines and so on. The energy consumption management is taken into account together with manufacturing efficiency. Energy is one of the major costs of automotive supply chain particularly the rubber part manufacturing. Most of the compression machines, injection machines are used the heaters to heat rubber materials and continuously heated whole day and whole weeks. It is sometimes worked around 20 days or more per month. Therefore, the energy is critical for the company in supply chine because cost is the major criterion of the competition. Energy control is dynamic and uncertain in aspects of obtaining the OEE of s firm. Scientific methods are applied to deal with the energy control. AHP and Fuzzy AHP including Fuzzy FMEA or even the combination of them. The criteria concern with the cause of uncertainty and risk such as severity, occurrence, and detection. The major goals of production control are cost control, improving

Cloud manufacturing system is emerged which is developed from the advanced manufacturing systems using the information and computer technologies such as cloud computing, internet of things (IOT), virtualization and service-oriented technologies, virtualization, cyber physical system. It transforms manufacturing resources and manufacturing capabilities into manufacturing services, which can be managed and operated in an intelligent and unified way to enable sharing facilities. The cloud manufacturing system can provide efficient and reliable, high quality, cost saving control. The system can enhance in the decision-making process while the material flow is tracked following the production process particularly at the final operation of the supply chain which needs to control on-time delivery. Problem solving can be rapidly done even immediately to increase productivity. The collaborative manufacturing management can be done along the whole life cycle of production. Few papers studied and presented could computing, cloud-based infrastructure. Lin Zhang et al. [30] expressed the cloud manufacturing which is the combination of cloud computing, IOT, serviced-oriented technologies and highperformance computing in order to solve the bottlenecks in the informatization development and manufacturing applications. It consists of three core components such as resources, cloud service and manufacturing on cloud. The evolutionary of modern manufacturing started from computer integrated manufacturing system (CIM). Then, the agile manufacturing is established to deal with time to market and supply chain management. It can support the responsiveness quickly while cost and quality can be consistently controlled. Afterward, concurrent engineering and collaborative design are developed to save time to market which was the life cycle of product taken into account. Peng Wang [31] explained the benefits and limitations of could computing for cloud manufacturing. Cloud computing is recognized at the moment that is an aggregate computing resources and a service of things. Presently, services on demand is dramatically increased when the world is in the digital disruption and transformation. Cloud computing is enabled to transition of computation in various forms such as platform, hardware, and software. The benefits of the cloud computing are to obtain lower start-up and operating costs, to ease of

access and scalability, to reduce risk on resource provision.

**216**

The characteristics of cloud management system are to envision for any feature of IOT such as self-service, visualization of system image, workload optimization,

management.

quality, and on-time delivery.

**3.4 Cloud management system**

#### **Figure 4.**

*The cycle of subcontractor in automotive supply chain management.*

method. It is not only cost reduction but also the quality improvement. The delivery plan is monitored by the SCM department of the 1st tier following with the QC inspection and packing progress in order to predict the efficiency of the supply chain management. All of the parts required on the exact due date are delivered to the 1st tier inventory which is controlled by RFID and barcode inside the centralized data based and linked to the OEM. The final step is to close the P&D order and evaluated the performance. The concept of CPOM, detailed in 4.3, system is applied to the 4 last steps of the SCM. First is the delivery plan. Then, the QC& packing work station and stock control. The 3rd operation is delivery parts to customers which is linked to the 1st tier inventory control. The last operation is to close the P&D order which is linked to evaluation system.

QC department at the 1st tier to be approved. The production is continuing if the first lot is confirmed to be acceptable parts. The finished goods are checked by

*Automotive Industrial Supply Chain Performance Evaluation under Uncertain Constraints…*

*DOI: http://dx.doi.org/10.5772/intechopen.93679*

This section presents the methodology of Neuro Fuzzy subcontractor evaluation before sending to the cloud production online monitoring (CPOM). This method is successfully validated and workable for automotive rubber subcontractor. The criteria include capacity, availability, quality, delivery, productivity, back order control. This study concerns only the monthly audit and yearly evaluation. As previous mentioned, the monthly is important because it affects the daily production performance. It is weight 70% of the full mark of the total evaluation procedure. Therefore, the yearly evaluation weight 30%. The subcontractors are

QC and packed before sending to the inventory, and close the P&D order.

**4.2 Neuro fuzzy subcontractor evaluation system**

*Production flow for supply chain collaborative of the 1st tier and subcontractor.*

**Figure 5.**

**219**

**Figure 5** shows production flow for supply chain collaborative of the 1st tier and the subcontractors. Based on the functional department. There are 5 departments which are defined in the same supply chain; engineering, planning supply chain, subcontractor and QC. The P&D order starts from the planning department and sends to the supply chain department and passes to the subcontractor. The engineering department prepares molds and operation standard procedure which is obtained from the new model testing at the real environment. The mold is sent to the subcontractor and tested for completion quality. Then, the subcontractor begins the preproduction and first lot experiment. The result is recorded and sent to the

*Automotive Industrial Supply Chain Performance Evaluation under Uncertain Constraints… DOI: http://dx.doi.org/10.5772/intechopen.93679*

**Figure 5.** *Production flow for supply chain collaborative of the 1st tier and subcontractor.*

QC department at the 1st tier to be approved. The production is continuing if the first lot is confirmed to be acceptable parts. The finished goods are checked by QC and packed before sending to the inventory, and close the P&D order.
