Efficiency Analysis of Die Attach Machines Using Overall Equipment Effectiveness Metrics and Failure Mode and Effects Analysis with an Ishikawa Diagram

Abstract

The semiconductor manufacturing sector has contributed to the advancement of technical development in the sphere of industrial applications, but one crucial factor that cannot be overlooked is the evaluation of a machine’s state. Despite the presence of advanced equipment, data on their performances are not properly reviewed, resulting in a variety of concerns such as high rejection rates, lower production output, manufacturing overhead cost issues, and customer complaints. This study’s goal is to evaluate the performance of die attach machines made by a prominent subcontractor semiconductor manufacturing business in the Philippines; our findings will provide other organizations with important insights into the appropriate diagnosis of productivity difficulties via productivity metrics analyses. The study focuses on a specific type of die attach machine, with machine 10 showing to be the most troublesome, with an overall equipment effectiveness (OEE) rating of 43.57%. The Failure Mode and Effects Analysis (FMEA) identified that the primary reasons for the issue were idling, small stoppages, and breakdown loss resulting from loosened screws in the work holder. The risk priority number (RPN) was calculated to be 392, with a severity level of 7, an occurrence level of 7, and a detection level of 8. The findings provide new insight into the methods that should be included in the production process to boost efficiency and better suit the expectations of customers in a highly competitive market.

1. Introduction

Every manufacturing firm aims to offer defect-free goods and exceed client expectations with impeccable service [1]. Products from the semiconductor manufacturing industry are utilized in a variety of applications, including commercial, consumer, industrial, and communication parts or devices [2]. Without a doubt, the standards of end consumers who would utilize goods including components from semiconductor manufacturing businesses make them selective and sensitive. At this time, the development and use of a higher model machine for semiconductor manufacturing firms continues, as each semiconductor manufacturing company seeks a competitive advantage over their market competitors. The adoption of higher machine models has increased in recent years [3]. Higher machine models are qualified as a result of quality-related concerns arising from customer feedback on shipped lots, with undetected defects and a high rejection rate occurring as a result of machine-related defects [4]. Additionally, machine flexibility is required to suit the process, customer, productivity, and automotive needs of clients. The ongoing enhancement of the features of the higher machine models, as well as improvements from previous models, has resulted in a minor improvements in quality-related issues noted in customer feedback.

It is important to highlight that the aim of this research is to effectively utilize the enhanced features of the newer models of the machines by optimizing their parameters. However, the optimization process is challenging due to the occurrence of machine malfunctions caused by mechanical parts, features, and other production-related factors. The possibility of reaching the “best” solution or optimization is determined by a number of factors, including the precise aims of the optimization process and a study’s environment. In this situation, it is necessary to identify the most optimal settings by maximizing the processing efficiency, which includes variables like limiting downtime, eliminating errors, and optimizing resource use. With this, it is important to aim to develop factors that will lead to increased overall processing efficiency. While no criterion ensures a perfect assurance of finding the optimum answer, the method focuses on improving parameters to increase efficiency.

The semiconductor industry is witnessing tremendous technical developments, particularly spurred on by Moore’s Law, which forecasts that the number of transistors on a chip will double every two years [5]. These innovations need the constant use of new manufacturing equipment in order to maintain performance gains and cost savings.

Recent advances, such as the combination of nanotechnology and 3D IC (integrated circuit) technologies, highlight the importance of cutting-edge research and development. Lau [6] emphasizes the value of 3D integration in semiconductor technology for improving device performance and functionality. Furthermore, as MOSFET devices approach their scaling limits, the industry adapts its business strategies to preserve a competitive edge. Saha [7] highlights how different segments are responding to next-generation technologies, transforming foundries into entire product development solutions, and implementing creative methods across fables and integrated device manufacturers. Incorporating these trends gives a thorough knowledge of the dynamic environment that influences the research, emphasizing the continuing technological and business model evolutions that drive equipment adoption and production efficiency.

In detail, the semiconductor industry is a crucial sector that propels technological progress and economic growth worldwide. To remain competitive, semiconductor manufacturers must constantly enhance their manufacturing efficiency and equipment performance. Overall equipment effectiveness (OEE) is a crucial statistic for evaluating and improving equipment performance. OEE is a well-known metric that measures equipment productivity across availability, performance, and quality.

Several studies have looked into the use of OEE in the semiconductor sector, highlighting its usefulness in detecting and resolving production inefficiencies. For example, Oechsner et al. [8] established the idea of overall fab effectiveness (OFE), which is an extension of OEE that is designed specifically for semiconductor fabrication. Their research showed that combining OEE with other performance indicators could provide a more complete picture of semiconductor manufacturing efficiency.

Chong and Ng [9] investigated the relationship between OEE, throughput, and production component cost in the semiconductor manufacturing business. Their research emphasized the relevance of OEE in streamlining production processes and lowering costs, ultimately improving overall operational efficiency. They gained vital insights into how OEE may drive cost-effective production by examining the interdependence of these variables.

Moreover, Giegling et al. [10] describe the implementation of an OEE system at a semiconductor company. Their findings demonstrated both the practical challenges and benefits of using OEE as a performance monitoring tool. The study emphasized the large increases in equipment utilization and industrial productivity that might be obtained through systematic OEE deployment.

The research builds on these basic studies by analyzing the efficiency of die attach machines in semiconductor manufacturing using OEE measurements, Failure Modes and Effects Analysis (FMEA), and Ishikawa diagrams. By doing so, the study aims to provide a thorough diagnosis of technical issues and recommend focused solutions. This study adds to the current body of information on OEE application in the semiconductor sector while also providing practical insights for improving equipment performance and productivity.

Indeed, the study intends to analyze quantitative indicators for gauging individual productivity using the approach OEE. The OEE comprises three key indicators: rate of performance, rate of availability, and rate of quality. These are linked with the six big losses. These losses are classified into six categories: breakdown, setup and adjustment, defect and rework, start-up, decreased speed, and idling and brief halt. These were identified by Nakajima in 1988 [11]. Additionally, the machine factor error can be detected by using a cause-and-effect diagram to assess the internal concerns related to the six significant losses, as suggested by Ishikawa [12]. Furthermore, the categorization of machine defects will be defined by the document of FMEA, centered on the risk priority number (RPN) findings [13]. The goal of this research is to conduct a thorough investigation of the productivity of die attach machines made by a semiconductor manufacturing company by calculating their overall equipment effectiveness (OEE) and identifying significant elements that contribute to machine inefficiencies. This investigation uses the six large losses framework, Failure Modes and Effects Analysis (FMEA), and the Ishikawa diagram to identify technical issues and enhance machine performance.

Die attach machine efficiency is critical to overall productivity in semiconductor manufacturing. Nevertheless, a thorough integration of techniques designed especially for semiconductor production environments is sometimes lacking in the published literature. The effectiveness of OEE and FMEA in evaluating production efficiency has been shown by earlier studies. Fundamental insights into the use of these metrics and failure analysis approaches in semiconductor production systems are provided by studies by Oechsner et al. [8], Chong and Ng [9], and Giegling et al. [10]. Various studies highlight how crucial it is to combine various approaches in order to improve operational effectiveness and pinpoint crucial areas for manufacturing process improvement. Nonetheless, the particular utilization of these techniques for semiconductor die attach machines is still inadequately investigated, underscoring the originality and significance of the present research.

Strict quality control must be maintained during semiconductor production in order to prevent customer complaints and guarantee business continuation. As previous research has shown, machine-related problems frequently result in differences between expected and actual output, which is why this work tackles these serious problems [14]. This research focuses on the die attach process, which is critical to semiconductor manufacturing [15]. It seeks to address persistent machine-related problems such mechanical problems and malfunctions and offer comprehensive insights into productivity metrics analysis. In order to detect and reduce productivity losses, the main goals of the study are to perform a thorough analysis of die attach machine productivity using Ishikawa diagrams, Failure Modes and Effects Analysis (FMEA), and overall equipment effectiveness (OEE). In order to clarify important variables impacting productivity, the study also compares the performance of several die attach machine types inside a semiconductor manufacturing company. Through comparison with earlier studies, especially that of Fam et al. [16], this research seeks to validate the applicability of OEE by adding fresh findings to the body of knowledge on productivity measures and OEE in semiconductor production. In keeping with accepted procedures described by Fam et al. [16], the project also intends to use total productive maintenance (TPM) techniques to promote the continual improvement of machine performances.

The main inquiry driving this investigation is as follows: what are the potential applications of Ishikawa diagrams in conjunction with overall equipment effectiveness (OEE) and Failure Modes and Effects Analysis (FMEA) to detect and reduce productivity losses in die attach machines in the semiconductor industry? Even while productivity measures are crucial for semiconductor manufacturing, there is still a big gap in the thorough examination of these metrics, especially when it comes to the die attach process. The merging of OEE, FMEA, and Ishikawa diagrams to offer a comprehensive knowledge of machine efficiency and productivity losses has not been fully investigated in prior research. In order to close this gap, this study will thoroughly evaluate the use of OEE in assessing die attach machines, use FMEA and Ishikawa diagrams to identify productivity loss factors, and compare the results with those of earlier studies to determine whether OEE is a useful metric for increasing equipment efficiency in the semiconductor industry.

The anticipated benefits of this research include increased profitability for companies that manufacture semiconductors by detecting and reducing productivity losses, as well as optimized efficiency through the combination of OEE, FMEA, and Ishikawa diagrams; this combination offers a thorough method for maximizing machine efficiency. Furthermore, it is expected that the results will raise the production facilities’ overall equipment effectiveness. By fulfilling these goals, the study hopes to advance the field of production and operations management by offering insightful analysis and useful suggestions for the semiconductor manufacturing sector. By filling a significant void in the existing literature, the integration of OEE, FMEA, and Ishikawa diagrams offers a fresh and thorough method for comprehending and enhancing productivity in die attach machines.

2. Methodological Approach

2.1. Gathering of Information and Data

Between August 2022 and December 2022, the assembly area of a well-known subcontractor in the semiconductor manufacturing industry in the Philippines was chosen as the site for the study. In this study, 7 days per week are considered as working days, with 3 shifts of 8 h each per day. Additionally, the layout of the machines uses a line layout or a product layout because this is optimized for mass production and the high-volume manufacturing of standardized products. The researcher utilized various data collection methods, including conducting interviews, conducting inspections, surveying the equipment, and analyzing past relevant data. The gathering of the necessary information and data focused on the assembly’s current issue, and past relevant data from the same time period was analyzed to evaluate the present issue. Furthermore, the study primarily focused on issues that impact product performance.

The machine type used in the study is called ASM8312 from ASM International, Almere, Netherlands. In detail, the ASM8312 is a die attach machine used in the front of line (FOL) semiconductor manufacturing process. It is primarily utilized in the epoxy die attach method, which uses epoxy adhesive to attach the die to the lead frame. This is a machine that is not fully automated; it has a parametric pick-up force, a bond force, and an epoxy dispensing adjustment function, as well as a semi-automatic wafer loader [17,18].

To evaluate machine efficiency and productivity losses, the researchers used a rigorous process that included overall equipment effectiveness (OEE), Failure Modes and Effects Analysis (FMEA), and Ishikawa diagrams. OEE offered a comprehensive statistic that combined performance, availability, and quality. This metric was essential for locating and fixing operational inefficiencies unique to the semiconductor manufacturing industry. Prioritizing risk mitigation techniques, FMEA methodically assessed possible failure modes and their effects with the aid of Ishikawa diagrams. In order to guarantee consistency and dependability in the study, data from 20 identical die attach machines that have been in operation since April 2017 were carefully collected. Comprehensive analyses of these data made it easier to extract important findings, which are explained in the sections that follow.

This research’s choice of case study methodology is crucial for a number of reasons. First off, the procedures involved in producing semiconductors are extremely intricate and context-specific, especially when die attach machines are involved. Through an in-depth examination of these intricacies in an actual situation, a case study design provides insights that would not be obtained using only quantitative or experimental methods. Second, this study intends to close a gap in the literature by concentrating on a particular semiconductor manufacturing company and its die attach machines. This research integrates OEE, FMEA, and Ishikawa diagrams in a context-specific and in-depth manner. This method advances the findings’ applicability in real-world situations and advances the theoretical knowledge of operational efficiency in semiconductor manufacturing.

2.2. Theoretical Framework

To identify the factors causing interruptions in the assembly of semiconductors, the study implemented a method of evaluating and assessing equipment performance. Moreover, the efficacy of the equipment was investigated using the OEE approach. The overall machine malfunction was measured to determine the equipment with the most significant equipment breakdown. The technique of six big losses was employed to categorize the losses and calculate them. An Ishikawa diagram was constructed after developing the types of causes using company data. The data from the Ishikawa diagram were analyzed, and FMEA computations were derived from past analytical methods. Subsequently, FMEA was used to determine the priority in addressing the factors to enhance effective performance. The study developed a theoretical framework, as illustrated in Figure 1, to better comprehend the process flow.

2.3. Equations

The study employed equations derived from Nakajima’s formula for equipment effectiveness to determine the OEE of the machine [11]. Before computing the OEE, the rates of quality, availability, and performance were calculated. To determine the six significant losses, the researcher executed calculations.

2.3.1. OEE Equations

The machine’s productivity performance was calculated using OEE in the research. In Equations (1)–(4), the formulas are drawn from the Nakajima [11] study.

$$AvailabilityRate=\frac{OperatingTime}{LoadingTime}\times 100\%$$

(1)

$$PerformanceRate=\frac{IdealCycleTime\times ProcessAmount}{ProcessingTime}\times 100\%$$

(2)

$$QualityRate=\frac{ProcessAmount-DefectAmount}{ProcessAmount}\times 100\%$$

(3)

$$OEE=AvailabilityRate\times PerformanceRate\times QualityRate$$

(4)

2.3.2. Six Big Losses Formulas

After computing the OEE, a technique known as the six big losses was utilized to identify the independent factors that should be removed from the assembly process [19]. Equations (5)–(10) show the six significant losses formulas that were employed in the study. The six big losses methodology was used to identify the factors that hindered the manufacturing process, following the calculation of OEE. Romanenko and Baybus [19] described this technique. Formulas for the six significant losses, as shown in Equation (5) through (10), were utilized in this research.

$$BreakdownLoss=\frac{BreakdownTime}{LoadingTime}\times 100\%$$

(5)

$$SetuporAdjustmentLosses=\frac{SetuporAdjustmentLosses}{LoadingTime}\times 100\%$$

(6)

$$IdlingandMinorStoppages=\frac{\left(TargetProduction-TotalProduction\right)}{LoadingTime}\times CycleTime\times 100\%$$

(7)

$$ReducedSpeed=\frac{(ActualCylceTime-IdealCycleTime)\times TotalProductProcessed}{LoadingTime}\times 100\%$$

(8)

$$DefectLosses=\frac{TotalReject\times IdealCycleTime}{LoadingTime}\times 100\%$$

(9)

$$ReducedYield=\frac{Scrap\times CycleTime}{LoadingTime}\times 100\%$$

(10)

Notes: Total rejects are finished goods that do not meet quality requirements, whereas scrap refers to useless materials or byproducts produced at any stage of the production process.

Cycle time is the total time taken to complete one cycle of a process or operation. This includes the time taken to perform all activities required to produce one unit of product, from start to finish.

Loading time is the amount of time that machinery is not working to its maximum capacity because it is waiting on parts, supplies, or personnel to load or prepare the machine for use.

3. Findings

3.1. Machine Malfunction

The study employed data from a notable semiconductor manufacturing firm in the Philippines. To identify the equipment that significantly affects the manufacturing process, the study gathered data on the downtime of each machine. Machine 10, which represents the primary production process challenges, was identified as the most problematic machine, as indicated in

Table 1. Downtime durations.

Equipment			Period			Sum of Malfunction
	August (min.)	September (min.)	October (min.)	November (min.)	December (min.)
1	1609.00	1851.00	1896.00	1921.00	1799.00	9076.00
2	1819.00	1862.00	1899.00	1743.00	1633.00	8956.00
3	1816.00	1979.00	1667.00	1707.00	1720.00	8889.00
4	1725.00	1563.00	1908.00	1871.00	1627.00	8694.00
5	1864.00	1679.00	1984.00	1555.00	1750.00	8832.00
6	1968.00	1555.00	1909.00	1586.00	1579.00	8597.00
7	1963.00	1920.00	1984.00	1567.00	1879.00	9313.00
8	1725.00	1662.00	1700.00	1763.00	1698.00	8548.00
9	1645.00	1616.00	1740.00	1762.00	1852.00	8615.00
10	1942.00	1958.00	1906.00	1854.00	3696.00	11,356.00
11	1626.00	1993.00	1630.00	1896.00	1623.00	8768.00
12	1740.00	1600.00	1917.00	1818.00	1883.00	8958.00
13	1553.00	1798.00	1651.00	1845.00	3684.00	10,531.00
14	1934.00	1623.00	1810.00	1972.00	1816.00	9155.00
15	1787.00	1935.00	1707.00	1992.00	1575.00	8996.00
16	1905.00	1770.00	1848.00	1751.00	1888.00	9121.00
17	1956.00	1638.00	1751.00	1888.00	1777.00	9010.00
18	1569.00	1673.00	1806.00	1788.00	1841.00	8677.00
19	1961.00	1802.00	1770.00	1895.00	1737.00	9165.00
20	1793.00	1689.00	1987.00	1694.00	1616.00	8779.00

. Hence, the study concentrated on analyzing machine 10 and gathered all the necessary information for its study. The gathered information and data were documented to compute the OEE score, which could aid in resolving the six major losses.

The purpose of collecting data from 20 machines, as indicated in Table 1, is to thoroughly investigate and determine which of these similar machines is the most troublesome. All 20 machines were of the same type, were acquired in April 2017 and have been functioning on the manufacturing line since then. They are all used to make the same sort of products. This consistency guarantees that the study is carried out under constant settings, allowing for a valid comparison. The inclusion of data from the other 19 machines in the main body of the paper is critical because it gives a solid foundation for emphasizing that, based on the statistics, machine 10 is the most troublesome. Given identical operational timelines and conditions, all machines experience the same amount of wear and tear. This technique not only validates the study’s findings, but also allows for a rapid and clear assessment of each machine’s relative performance.

3.2. Overall Equipment Effectiveness

The average OEE value computed from Equations (1)–(4) is presented in

Table 2. Computation of overall equipment effectiveness.

Period	Percentage of Availability (%)	Percentage of Performance (%)	Percentage of Quality (%)	Overall Equipment Effectiveness (%)
August	78.51	68.24	84.90	45.49
September	79.61	64.75	84.86	43.74
October	82.61	61.61	84.87	43.20
November	80.44	67.64	84.89	46.19
December	76.01	60.85	84.88	39.26
Average	79.44	64.62	84.88	43.57

. The calculated overall equipment effectiveness score of the machine was found to be noticeably lower than the international standard of 85%, according to Romanenko and Baybus [19], indicating that the manufacturing process requires improvement. Nevertheless, the combined criterion for world-class overall equipment effectiveness (OEE) is usually set to 85% [19]. This benchmark is the result of combining the various benchmarks for OEE’s three essential components: availability, performance, and quality. World-class OEE is obtained by multiplying the benchmark percentages of availability (90%), performance (95%), and quality (99%) [19]. When these percentages are combined—0.90 for availability, 0.95 for performance, and 0.99 for quality—the result is around 85% [19]. This computation emphasizes the rigorous criteria needed to attain world-class OEE, which reflects maximum efficiency and minimal losses in a manufacturing operation [19].

The world-class OEE is a benchmark rather than a global standard, and it is frequently only achievable in large-scale production contexts. It establishes high standards for equipment efficiency and performance, frequently requiring maximum utilization and minimal downtime. While not a universal standard, world-class OEE signifies exceptional operational excellence and is sought as a goal in high-volume manufacturing environments. Still, identifying the underlying issue is critical to resolving the problem. To categorize the mistakes that impact the performance of machine 10, the researchers calculated the six big losses.

3.3. Six Big Losses

According to the calculated outcomes, the major loss was attributed to the idling and minor stoppages, succeeded by the breakdown or loss, reduced speed, defect losses, reduced yield, and adjustment or set-up losses. The corresponding mean values for these losses were 21.92, 16.55, 13.47, 9.74, 5.38, and 4.02, as presented in

Table 3. Computation for the big losses.

Month	Idling and Minor Stoppages (%)	Breakdown Loss (%)	Reduced Speed (%)	Defect Losses (%)	Reduced Yield (%)	Adjustment or Set-Up Losses (%)
August	20.08	17.27	11.68	9.54	5.56	4.22
September	21.01	16.49	14.24	9.82	5.32	3.9
October	22.55	13.09	15.84	9.65	5.48	4.3
November	20.13	15.63	12.23	9.99	5.12	3.93
December	25.81	20.25	13.34	9.69	5.43	3.74
Average	21.92	16.55	13.47	9.74	5.38	4.02

.

3.4. Fishbone Analysis

The fishbone diagram study, as shown in Figure 2, contains the top idle and minor stoppages and breakdown loss found in the company’s historical data, since this will be the focus of the investigation. The bond head assembly, dispensing assembly, work holder assembly, and die ejector assembly are all included in this category of die attach machine assemblies. The classification of categories was made utilizing historical data from the machine’s error logs and preventive maintenance records, which are linked to the machine parameters investigated in this study.

3.5. Machine Errors

Table 4. Machine errors summary.

Jam on Lead Frame Push Out Rolls			Contaminated Pick-Up Tool		Wafer Centering and Measurement Failed		Loosened Work Holder Screws		Die Ejector in Collision with Wafer Table
Month	Repair Time (min)	Amount of Damage (no.)	Repair Time (min)	Amount of Damage (no.)	Repair Time (min)	Amount of Damage (no.)	Repair Time (min)	Amount of Damage (no.)	Repair Time (min)	Amount of Damage (no.)
August	3000	10	4000	7	1000	1	25,000	20	1100	1
September	1500	5	2000	3	0	0	12,500	10	0	0
October	2500	10	5000	12	3100	3	14,000	10	2200	1
November	0	0	7000	20	2700	2	12,000	5	0	0
December	2000	10	0	0	1400	1	16,000	10	900	1
Total	9000	35	18,000	42	8200	7	79,500	55	4200	3

displays the top five equipment errors identified in examining the connection between machine errors and minor and idling halts, breakdown/malfunction loss, and the Ishikawa diagram for machine 10 only. Machine 10’s data were chosen because this machine is the most problematic based on the data gathered. Machine defects include jams on lead frame push out rollers, contaminated pick-up tools, failed wafer centering and measurement, loosened work holder screws, and die ejector colliding with wafer table, as indicated in Table 4. Furthermore, according to the study, loosened work holder screws occur the most frequently and require the most time to fix, followed by contaminated pick-up tool, the jam on the lead frame push out rolls, failed wafer centering and measurement, and die ejector collision with wafer table. The chart also displays the overall repair time and damage for the previous five months.

3.6. The Failure Mode and Effect Analysis

Results coming from failure mode and effect analysis can be determined using the RPN calculation output (RPN is a product of detection, occurrence, and severity), as shown in

Table 5. FMEA.

Process	Failure Mode	Effect of Failure	Severity	Causes	Occurrence	Current Controls	Detection	RPN	Recommended Action
Die Attach Process	The machine cannot continue processing	When manufacturing time is halted, the number of good items decreases, requalification and immediate actions decreases machine productivity	7	Jam on lead frame push out rolls	5	Use of PM for the checking process	2	70	Inclusion of checking of the condition of the push out rolls during the set-up in the work instruction
				Contaminated pick-up tool	5	Use of a set-up checklist	7	245	Inclusion of checking of the pick-up tool at the stage of process monitoring every two hours in the work instruction
				Wafer centering and measurement failed	5	Work instruction to perform three-point alignment	2	70	Inclusion of checking of wafer centering every wafer change in the work instruction
				Loosened work holder screws	7	Visual inspection on work holder every PM	8	392	Inclusion of work holder plate screw condition on set-up checklist
									Inspection and testing of work holder planarity (1st—piece check)
				Die ejector in collision with wafer table	4	Use of a set-up checklist	2	56	Inclusion of checking of the die eject needle at every wafer change in the work instruction

. The RPN value determines the severity of the situation, with a higher RPN indicating a more serious issue. The most crucial error that needs to be resolved urgently is the work holder screws becoming loose, with a high RPN score of three hundred and ninety-two. Following that, other errors include contamination of the pick-up tool, jams on lead frame push out rolls, failure in wafer centering and measurement, and die ejector colliding with wafer table, with RPN scores of two hundred and forty-five, seventy, seventy, and fifty-six, respectively. The failure mode and effect analysis involved gathering input from various sources, such as the affected operators, equipment experts, maintenance engineers, preventive maintenance engineers, production supervisors, quality engineers, and process engineers. The table also includes the current issue controls and recommended steps based on interviews with the people directly involved in the activities conducted by the researchers.

4. Interpretation

According to the findings of the study, the productivity metrics analysis is a significant approach for semiconductor manufacturing businesses, especially when dealing with equipment productivity. OEE measures played a crucial role in determining whether the equipment is working as intended. According to Hedman et al. [20], it is the responsibility of a company to understand the key elements of OEE, which include availability, performance, and quality rate. These elements play a critical role in identifying the underlying issues affecting the system.

The study discovered that machine 10 obtained an availability/operability rating of 79.44%, a performance/productivity rating of 64.62%, and a quality/scrap rating of 84.88%; these all fall below the world-class benchmark. The findings suggest that the performance parameter is the weakest among the three, indicating a possible mechanical malfunction in the manufacturing equipment. The levels of availability and quality are superior to that of the performance, yet they are still lower than the predetermined standards. Nevertheless, machine 10 emerges as a vital component in the manufacturing process, acting as a bottleneck due to its large impact on total production throughput. The analysis shows that the efficiency of machine 10 has a direct impact on processing speed, making it critical for improving production efficiency and throughput. The selection of machine 10 as the most troublesome underscores the critical need for targeted modifications to eliminate bottlenecks and improve overall production performance. Furthermore, the root cause analysis performed on machine 10 provides useful information about potential issues that may develop in other machines. By addressing these root causes across all machines, preventive steps can be introduced to reduce risks and prevent similar inefficiencies, improving overall production efficiency and dependability. This addition stresses the necessity of addressing difficulties with machine 10 and the proactive approach required to prevent similar problems from affecting other machines in the manufacturing process.

Moreover, it is possible that the firm falls short in regularly assessing the overall equipment effectiveness of its machines, as seen not only in the examined machine but also in the others. Past studies indicate that achieving a 100% quality rate may be unrealistic in some situations, but the machine selected in the current study falls well below the 99% benchmark. Additionally, Kuhlang et al. [21] discovered that several companies lack a fundamental understanding of analyzing productivity metrics and the concept of the theoretical highest level of performance, as demonstrated by the data presented. This implies that the company does not routinely compare its performance to the established standards.

The major causes of the six major losses were identified; idling and minor stoppages were succeeded by breakdown or loss, reduced speed, defect losses, reduced yield, and adjustment or set-up losses. The corresponding mean values for these losses were 21.92, 16.55, 13.47, 9.74, 5.38, and 4.02, respectively. Among these, idling, minor stoppages, and breakdown loss were found to be the most significant contributors, and further analyses were carried out on these areas. The researchers extensively examined the machine error report to determine the underlying root of the issue, and the findings revealed that loosened work holder screws had the highest repair time of 79,500 min and the highest damage frequency at 55. The remaining issues include a contaminated pick-up tool, a jam on the lead frame push out rolls, a failed wafer centering and measurement, and a die ejector collision with the wafer table. The FMEA, based on interviews with manufacturing process participants and industry experts, was the last component of the productivity metrics analysis. Loosened work holder screws had the highest RPN of 392, followed by a contaminated pick-up tool, a jam on the lead frame push out rollers, a failed wafer centering and measurement, and a die ejector collision with the wafer table.

The findings are summarized in

Table 6. Summary of findings.

Metrics for Measuring Productivity	Results
1. Overall equipment effectiveness (OEE)	This indicates that the overall equipment effectiveness of the firm was 43.57%, with rate of availability, rate of performance, and rate of quality of 79.44%, 64.62%, and 84.88%, respectively. These rates are below the standard of excellence that is set at 85% globally.
2. Six big loses	Idling and small stoppages (21.92%) contributed the most to the losses, followed by the breakdown loss (16.55%).
3. FMEA	The loosened work holder screws were found to have the biggest risk priority number with three hundred ninety-two. The other evidence of overall equipment effectiveness losses included a contaminated pick-up tool, a jam on lead frame push out rollers, a failed wafer centering and measurement, and a die ejector collision with a wafer table.

, which adequately fulfills the need for a succinct review of the data while minimizing redundancy and enhancing clarity. To further prevent confusion, the researchers made sure that all of the names and numerical data are listed consistently throughout the manuscript. This thorough report validates the conclusion that machine 10 is the most troublesome, highlighting the necessity of targeted maintenance and investigation.

While the study has produced significant results, there are some limitations that should be acknowledged. One of these limitations is the small sample size used for the machine under study; including more machines in the research could potentially enhance the OEE outcomes. In addition, due to the time constraints of the data collection process, the study only focused on the machine aspects, and other factors related to assembly line operations, setting adjustment techniques, and more assembly issues that might yield substantial effects were not explored.

In the future, it would be beneficial to increase the number of machines tested to obtain more comprehensive results. Additionally, other methods like the Taguchi method, which has been used in various industries to enhance process parameters, should be implemented to assist in system development, parameter specification, and tolerance analysis for assembly line systems. Previous studies have shown the effectiveness of this method [22,23], which could improve the quality of data and provide more depth to the analysis. This approach could be explored in future research to expand upon the current study.

By critically evaluating current theories and studies to pinpoint specific research gaps, defending the methodologies employed, adding more references and specific statements to the literature review, articulating research objectives and hypotheses with clarity, and carrying out a more thorough productivity analysis, putting these approaches into practice would enhance the research. By putting these approaches into practice, one can ensure that the research contributes more significantly and in a way that is both scientifically and practically useful to the field of operations management. This can be carried out by conducting a more thorough productivity analysis, expanding the literature review with more specific statements and references, defining research objectives and hypotheses clearly, and critically analyzing current theories and studies to identify specific research gaps. These enhancements will guarantee that the study contributes more significantly and in a way that is beneficial to science on the subject of operations management.

5. Conclusions

This study has shown how well die attach machine efficiency can be evaluated and increased in semiconductor manufacturing by combining OEE, FMEA, and Ishikawa diagrams. The utilization of the case study approach yielded significant insights into particular operational difficulties and potential areas of improvement within the manufacturing environment under investigation. Subsequent investigations may build on these results by utilizing analogous approaches in various semiconductor production scenarios or by looking at other factors that affect equipment efficiency. As such, more progress may be achieved in developing the theoretical understanding of manufacturing efficiency as well as its actual implementation.

The findings are congruent with those of Fam et al. [16], who showed the effectiveness of OEE and TPM methods in the electronic components and boards business. This study emphasizes the necessity of taking a systematic approach to equipment maintenance and performance monitoring, which can result in large productivity increases. Fam et al.’s integration [16] of OEE with TPM procedures provides a solid foundation for meeting world-class OEE benchmarks, especially in complicated production contexts [19].

The study emphasized the key factors that have a significant influence on machine efficiency, particularly engine errors, and the importance of using OEE measurements to detect and identify losses in the manufacturing process. The research indicates that, when evaluating machine efficiency, companies and managers should not distance themselves from operating sophisticated machinery. A lack of coordination between production and maintenance can adversely affect the maintenance processes used for these machines.

The research discovered that, out of the 20 die attach machines investigated, machine 10 had the most frequent malfunctions. Its availability, performance, and quality rates were found to fall below the global standards of 90%, 95%, and 99%, respectively, resulting in a low OEE value of 43.57%, which is far below the global requirement of 85%. The primary causes of loss were idling, minor stoppages, and breakdowns [19]. The FMEA and fishbone diagram findings revealed that the loosened work holder screws were the most critical issue, with an RPN of 392. The study’s results may be valuable to the semiconductor manufacturing sector in the Philippines and other similar countries. Manufacturing businesses may utilize the study’s methodology and findings to improve productivity and the importance of overall equipment effectiveness and the failure mode and effect analysis for productivity metrics analysis.

Furthermore, this paper presents an integration of FMEA, Ishikawa diagrams, and OEE to provide a thorough analysis of productivity losses and machine efficiency in the semiconductor sector. The results provide insightful information that practitioners can put to use, especially in identifying particular loss categories and their underlying causes. The modifications implemented in reaction to the reviewer’s input have improved the manuscript’s scientific rigor and clarity, guaranteeing that it makes a significant addition to the field.

Future research should look into the long-term impact of these enhancements, as well as other productivity metrics, to fine-tune the assessment of machine performance. By building on the approaches and insights given in this study, the semiconductor sector may achieve greater operational excellence while maintaining its essential position in the global economy.