Journal of Engineering Research and Reports

Asset integrity is closely intertwined with process safety where the latter is often perceived to be equivalent or a subset of the former. In Malaysia, the requirements for offshore process safety are set by Petronas assuming exclusive rights to petroleum in the nation. It imposes and enforces these requirements on oil and gas companies entering into its production sharing contracts via the common law. Process safety management in Malaysia is strongly influenced by the US OSHA 3132 with elements comprising process safety information, process hazard analysis, operating procedures, employee participation, training, contractors, pre-startup safety review, mechanical integrity, hot work permit, management of change, incident investigation, emergency planning and response as well as compliance audits. These elements are largely included in the Mandatory Control Framework of Petronas and the trio of design, technical and operating integrity adopted in the process safety management of other oil and gas companies. These management practices align with the reiterative plan-do-check-act model. Process safety performance is also gauged with indicators suggested by international institutions such as the American Petroleum Institute. On top of the Control of Industrial Major Accident Hazards Regulations 1996 for onshore processes, this study deems that establishing statutory law for offshore installations will be beneficial to propel offshore safety in Malaysia to a greater height.

This paper discusses the characterization of an integrated circuit (IC) silicon die fracture strength to have a realistic die crack assessment. The evaluation was conducted using a 3-point bend test setup to measure the die strength of actual IC dies. Both the active side and the back side of the IC die were tested for 2 types of dies with different active side circuit layout. Results showed that the difference in the die active side circuit layout or structure has impact on die strength. It was also found that the active side was weaker than the back side. This implies that both the active side and the back side of an actual IC die must be subjected to fracture strength characterization to have an assessment that would be in a better agreement with real condition. Using only the strength of the back side would result in over-estimating the die strength. The common approach of using the fracture strength of the die back side to characterize the die strength is not realistic and can mislead the assessment of die crack or semiconductor package robustness.

Interface adhesion tests are conducted to obtain a quantitative value of the adhesion strength or interface fracture toughness to forecast initiation and propagation of interface delamination failures in semiconductor packages. One of the common methods of fracture mechanics interface characterization is the four-point bending test. Problems with the experimental setup would result in having inaccurate results. In this study, different issues with the four-point bending test using mold/copper bi-material beam with notch were addressed. It was shown that incorrect anvil alignment and centering could give inaccurate fracture toughness results. The distance between anvil supports and the depth of the notch is also very important. From the study, it can be concluded that issues with experimental setup and the test sample must be addressed to ensure accurate results from four-point bending interface characterization.

Intermetallic is a critical factor that is controlled and limit during manufacturing of integrated units. Through improving the ball flatness response of the semiconductor wire significantly impact the outcome for the intermetallic characteristic between metal such as Gold-Aluminum connection. The implementation of this study on the actual manufacturing control eliminates escapee of poor intermetallic that would impact the reliability and integrity of the package upon its application to the end product.

This study presents three different mathematical models for profit optimization of agricultural products in Bangladesh. The prime focus of the paper has been to develop a Mixed Integer Linear Programming (MILP) model and analyze this model for two situation of demand uncertainty. Considering demand will be known before and after production. For the mentions of above two cases, we investigate the change of solution applying least demand, maximum perhaps demand and extreme demand scenarios. I think this is real life problem and this analysis will be helpful for all types of agricultural producers.  The proposed MILP model is to maximize the total profit and also to estimate the profitable production locations. The formulated MILP model were solved by A Mathematical Programming Language (AMPL) and results obtained by appropriate solver MINOS. Numerical example with the sensitivity of several parameters has been deployed to validate the models. Results show that maximum perhaps demand scenario gets better solution according to our expected value compare of other two scenarios.

The trend for use of rooftop solar photovoltaics (PV) is rising due to their promising economic potential as a source of clean renewable energy. In general, a source of renewable solar energy consists of solar PV, an automatic charge controller, a battery pack, and an inverter. The reliability of a rooftop solar PV system is evaluated herein as that of a coherent threshold system (CTS). First, we utilize the unit-gap method and the fair-power method to verify that a given Boolean function is a threshold one and to identify its threshold and component weights. Both methods utilize specific features of the Karnaugh map (K-map). The unit-gap method uses the map to list all necessary inequalities by inspection, and then reduce them significantly by omitting dominated ones. The fair-power method uses the Karnaugh map to compute Banzhaf indices by appropriate map folding followed by XORing of true cells and false cells.  We evaluate the CTS reliability via a recursive algorithm based on the Boole-Shannon’s expansion in the switching domain, which is transformed via the real transform to the total probability law in the probability domain.

This paper will discuss how package voids Mold defect was addressed on thinner Compression mold packages like BGA (Ball Grid Array) and Sensor Devices. Significant action was the application of “Rey Resin Dispense” pattern in compression mold granule compound dispensing was a key action in reducing package voids defect. Using a DMAIC methodology (Define, Measure, Analyze, Improve and Control), a powerful analytical tool that serves as a guide towards the success of eliminating package voids defect. Using this methodology, all Key Process Input Variables (KPIVs) of compression mold and its molding compound material were identified as X’s or potential causes and test its significance to affect Y-response which is the package voids. Funneling of X’s further trimmed to the most possible contributor to package voids. The remaining possible contributor X’s undergo statistical validation which point to granule compound resin dispensing as the main contributor to induce package voids defect. Several actions were tried but failed to zero-out the defect. With the help of brainstorming and imaginative ideas and data gathered on how to improve granule resin dispense of compression mold positive outcome was realized. From the concept of light ray, to naming it as “Rey Resin Dispense” significantly reduce package voids defect in all thin BGA and Sensor devices. This learning was then applied to all existing thin packages and to new packages in compression molding process.

Aeronautical navigation systems installed at specific locations within Saudi Arabia report the actual observed data of the times to failure (TTFs) for many types of engineering elements made by some typical international highly-acclaimed manufacturers. There is a need to compare these observed data to the estimated data provided by the manufacturers themselves. The paper gathers a lot of data on the actual failure instances of many components produced by a variety of manufacturers and installed at different locations within the extensive area of Saudi Arabia. These data are used to calculate the mean times between failures for these components. The paper points to an inadvertent discrepancy between these data and the corresponding mean times between failures (MTBFs) suggested by some prominent manufacturers. Such suggested MTBFs are typically optimistic and unrealistically high irrespective of the elements, the location, and the manufactures. The work reported herein is a preliminary assessment of this phenomenon that might lead to its theoretical modeling and subsequent understanding.

This paper discusses about direct material wastage brought by expiring molding compound particularly for QFN-mr Packages.

An Engineering problem solving methodology was used to identify the material, its consumption, cost, the root cause, and solution of the problem. The study aims to solve huge amount of compound material disposed in the manufacturing line.

By challenging the current 24hours suppliers thawing time, it is shown that at 16hours thawing time, molding compound already reached the required room temperature (23°C ± 3°C). To maximize the usage window of molding compound, the existing floor life of 24hrs was also reviewed and after experiment, it was found out that it can still be extended to another 24hrs or 48hrs total floor lifetime.

Combining the 2 improvements, a total of 56hrs usage window for the molding compound from 24hrs usage window.

Hydrological response of a catchment is a function of rainfall as influenced by catchment characteristics comprising geomorphology, land cover, and management practices. In this study, the analysis mainly focused on how geomorphological characteristics influence the catchment hydrological response. Geomorphological analyses of catchment geometry, stream patterns, relief, and slope can be used to characterize the catchment features that affect the drainage network. These characteristics are catchment specific and therefore unique to provide an insight into its hydrologic response. The objective of this research was to quantitatively analyze geomorphologic characteristics; linear, areal, drainage pattern, and relief aspect, of Amala River catchment, using ArcGIS tools and infer its hydrological behavior. The morphometry of the catchment was derived from the DEM within the ArcMap environment. These parameters as well as mathematical map equations were used to derive geomorphological characteristics such as bifurcation ratio, rho coefficient, drainage density, infiltration number, form factor among others. The results show that the Amala River catchment is elongated with uniform lithology and a higher probability of delayed peak hydrographs due to longer lag time and time of concentration. The catchment exhibits a dendritic drainage pattern with an average bifurcation ratio of 4.26 which is closer to the upper bound value of 5. This indicates a reduction in peak flows and a delayed time to peak. The surface runoff yield efficiency was low and non-uniform with an average drainage density of 1.073 km/km2. The catchment was characterized by higher infiltration characteristics as compared to surface flows, this varied spatially, with sub-basins far North of the outlet having high infiltration than those near the outlet. The catchment relief was characterized as steep and therefore high stream velocity was inferred. The investigation and findings of this study on catchment geomorphology and inferred hydrologic behavior will be of great importance in catchment management, water resource planning within the catchment, and water harvesting at a spatial scale. Thus, the outcomes provide a baseline for informed water pan and water harvesting structures site.

Pick-and-place (PnP) rubber tip is used for the die attach process in semiconductor packaging. During the die attach process involving a very thin integrated circuit (IC) die, voids in the die attach film (DAF) adhesive were observed. This paper presents the modeling study conducted to understand the impact of PnP rubber tip design on the occurrence of DAF voids. Mechanical modeling was done to simulate the deformation of the very thin die when it is picked using the PnP rubber tip. Different PnP rubber tip designs were modeled. The results revealed that the die deforms during the die pick up process because of the vacuum force, and rubber tip design affects the amount of deformation. It was also found out that more DAF voids occur when the resulting die deformation is larger.

In semiconductor packaging industry, silhouetted die defect is occasionally encountered in singulation process esp. on substrate land grid array (LGA) package with tight clearances. This paper is focused on the prevention of the silhouetted die occurrence as it would affect the assembly yield performance of the device. The silhouetted die is caused by the tight clearance of die edge to package edge, given the machine and process tolerances. Process enhancement and optimization were done through adjusting the die placement accordingly as per the defined measurement. Eventually, the occurrence of silhouetted die was successfully mitigated by formulating the appropriate die placement references.

Very thin semiconductor package is very prone to package crack. This paper discusses the stress modeling study conducted to understand the package crack problem in a specific smart card package. Finite element analysis (FEA) was used to analyze the maximum package stress level and corresponding location to find out if the presence of debris during the package assembly punching process could cause such problem and how it would happen. Based on the stress results, it was confirmed that even with a 60μm-thick piece of debris under the package, crack at the top is possible due to package bending and mold stress exceeding the flexural strength of the package mold material. The stress increases as the debris location is moved closer to the area where force is applied during the punching process. The study shows that the presence of debris should not be taken for granted though how small the debris may seem because significantly high bending stress could still be induced especially for very thin packages. Eliminating any source of debris in the package assembly process.is very important to prevent package crack.

This paper will discuss the improvement made to increase machine capacity in order to be prepare for incoming demand ramp-up and to free up machine allocation to back-up automotive machine. Using DMAIC methodology (Define, Measure, Analyze, Improve and Control) approach, mold curing time identified as the top contributor during time study of the process with 120 secs consumed time per shot. A Differential Scanning Calorimetry (DSC) and Curability Curve study for molding compound was conducted from both internal and external expertise to get the optimum lower range of cure time. Considering all the quality risk using Risk assessment, the success of reducing Mold curing time from 120secs to 80secs increases 25% machine capacity of transfer Molding for Quad Flat No leads(QFN) / Quad Flat No leads multi-Row(QFN-mR) with significant cost savings. The project was able also to free up 3 molding system to support upcoming products.

Package delamination is one of the problems in semiconductor packaging. Understanding the delamination mechanism in a specific situation is very important to identify the root cause and implement robust solution. In this study, package deformation modeling was done to analyze the deformation of the substrate or package at different thermal conditions. The modeling result was compared with the actual package deformation of the package with delamination problem. It was found out that the observed deformation through actual cross-section analysis matched with the modeling result at reflow temperature condition. Thus, it could be concluded that the delamination happens during package reflow and not after post mold cure or the preceding processes.