Search results

During the past few years, statistical process control and experiment design concepts have taken a prominent place within the industry. The use of such tools within the Motorola, GEG manufacturing environment, has grown to the point where reflow and wave solder process development and optimisation has significantly benefited. The ability to evaluate statistically and model various known and unknown phenomena has provided GEG's manufacturing technology with a series of very powerful tools to aid in process control and development. The primary purpose of this paper is to present the various approaches used by GEG to implement the previously mentioned statistical tools, with respect to the development of infra‐red (I‐R) reflow solder processes and enhancement of certain quality characteristics associated with wave soldered printed wiring boards (PWBs). Beyond specific GEG applications, the paper discusses the role of statistically designed experiments and process control methods as a vehicle for providing answers to complex manufacturing problems. In addition, a discussion of the mathematical and graphical methods underlying the interpretation of quantitative data is presented. Perhaps the most important benefit derived from the use of statistics to solve manufacturing and quality problems is related to decision making. When experiments are conducted to isolate unwanted sources of process and product variation, decisions must be made to determine whether or not certain experimental effects are important. Through the application of statistics, the researcher can ascertain the mathematical probability associated with the random chance occurrence of various experimental effects. With this knowledge, the researcher can make decisions with known degrees of risk and confidence. Without such knowledge, an organisation might possibly expend valuable resources and derive no direct benefit. Ultimately, the principal reason for applying statistical methods and procedures is to increase quality and yield, while simultaneously reducing costs.

A new product design required the addition of a second layer of electronics to control a base module. This product was designed with significant overhangs of heavy leads and components which presented a significant challenge to many different solder assembly processes. Only the heated gas jet process was able to solder the product successfully without damaging the printed wiring boards.

To answer the challenge, a new machine was developed, combining dispensing of solder paste with hot gas jet reflow technology. This provided a combination of capabilities resulting in a flexible process which was significantly superior to alternative technologies.

Other soldering processes such as laser, focused xenon lamp, robotic soldering iron, and focused IR soldering technologies were evaluated. Each of these technologies causes some damage or defect to the assembly due to the heat sinking aspect of the circuit assembled. These alternative processes would create damage or defects to the assemblies by burning the laminate, delaminating the pads on the printed wiring board, or not soldering the pads.

Proof of concept tests before machine designs were initiated demonstrated the potential and capabilities of this technology for automated assembly soldering. Testing indicated that the heated gas jet processing would provide a means of soldering the assemblies at a controllable rate without damaging the circuit boards.

While evaluating the machine ion its design phase, a designed experiment was initiated to help understand the relationships between head temperature settings versus gas flow rates, the measurable output was time to reflow.

The process meets all expectations in terms of solder fillet appearance, volume, and overall visual quality while maintaining process cycle time requirements.

Drawbridging or Stonehenge Effect of leadless components (i.e., the standing up on their end faces) has been investigated. An explanation is offered based on a straightforward theoretical model considering surface tension, sustained by a great amount of experimental evidence. The phenomenon is strongly associated with condensation reflow soldering, whereas the dimensions of the metallisation at the underside of the leadless components and the size of the solder lands on the board are major influencing factors. Practical hints are given to overcome the problem.

A mechanical and microstructural study was performed of 43/43/14 tin/lead/bismuth solder. This alloy melts lower than the commonly used tin/lead solders and therefore holds promise as a useful material in two‐step soldering processes or in processes with thermally sensitive components. Mechanical testing of 43/43/14 tin/lead/bismuth showed a strength comparable to that of tin/lead solders but increased creep rate. Microstructural analysis (scanning electron microscopy/energy dispersive X‐ray) exhibited the same mechanism of fatigue as for tin/lead solders, viz., heterogeneous coarsening. Thermocyclic fatigue demonstrated that the long‐term reliability of 43/43/14 tin/lead/bismuth is comparable to that of tin/lead solders.

This study aims to apply for the first time in literature a new multi-objective sensor selection and placement optimization methodology based on the multi-objective Lichtenberg algorithm and test the sensors' configuration found in a delamination identification case study.

This work aims to study the damage identification in an aircraft wing using the Lichtenberg and multi-objective Lichtenberg algorithms. The former is used to identify damages, while the last is associated with feature selection techniques to perform the first sensor placement optimization (SPO) methodology with variable sensor number. It is applied aiming for the largest amount of information about using the most used modal metrics in the literature and the smallest sensor number at the same time.

The proposed method was not only able to find a sensor configuration for each sensor number and modal metric but also found one that had full accuracy in identifying delamination location and severity considering triaxial modal displacements and minimal sensor number for all wing sections.

This study demonstrates for the first time in the literature how the most used modal metrics vary with the sensor number for an aircraft wing using a new multi-objective sensor selection and placement optimization methodology based on the multi-objective Lichtenberg algorithm.

We introduce a risk management framework to assess food security, which is interpreted as the probability of fatality or adverse health effects due to lack of food and which is a product of food availability, access, and vulnerability.

We derive cost-minimizing policies to achieve food security objectives by addressing availability, access, and vulnerability, and taking into account how randomness, uncertainty, and heterogeneity affect the system.

Ignoring key sources of variability, particularly heterogeneity, may lead to biases because food security policies require targeting the most vulnerable populations, which may each have unique features such as age, location, and health status. Establishing any policy solution requires making tough choices about policy criteria. Outcomes will differ when the criteria is to minimize overall risk or to minimize risk to the most vulnerable.

Policies addressing food security crises should balance enhanced supply with targeting available food and the provision of emergency health services to vulnerable populations.

This study examines how foreign R&D investment may explain interfirm variations in productivity performance of home country firms in terms of spillovers. Many have studied spillovers from MNCs to host country’s firms, but there is still scarce evidence on spillovers from outward FDI to the home country. This study analyzes spillovers from foreign R&D investment and hypothesizes that the benefit of outward R&D spillovers occurs only when knowledge accumulated in foreign R&D centers is effectively transferred to MNCs’ parent companies at home. This benefit depends on the mandate of foreign R&D units, their embeddedness in the host economy, and their entry mode. Using detailed firm-level data for Switzerland, our findings seem to support our arguments.

This chapter aims at providing an understanding of the research and devlopment (R&D) process in the pharmaceutical industry, by exploring the methodological challenges and approaches in the assessment of the determinants of innovation in the pharmaceutical industry. It (i) discusses possible methodological approaches to model occurrence of events; (ii) describes in detail competing risks duration models as the best methodological option in light of the nature of pharmaceutical R&D processes and data; (iii) concludes with an estimation strategy and overview of potential covariates that have been found to correlate with the likelihood of failure of R&D pharmaceutical projects.

This study aims to develop a numerical identification and characterization of crack propagation through the use of a new optimization metaheuristics called Lichtenberg optimization.

The damage-identification problem is treated as an inverse problem, which combines finite element methods with intelligent computational methods to obtain the best possible response. To optimize the objectives, the Lichtenberg algorithm is applied, which includes concepts of random cluster growth in nature.

The simulations show that it is possible to determine the Lichtenberg spectrum algorithm a part of the structure to be removed and replaced in this case to stop the propagation.

The results show a very good crack identification in plates-like structures using the Lichtenberg algorithm (LA) based only in strain fields. Although many studies have reported on damage-identification-based optimization methods, very few have focused on the crack tip modeling and LA as the main solver.

The present paper aims at the multi-objective optimization of a reentrant hexagonal cell auxetic structure. In addition, a parametric analysis will be carried out to verify how each of the design factors impact each of the responses.

The multi-objective optimization of five different responses of an auxetic model was considered: mass, critical buckling load under compression effort, natural frequency, Poisson's ratio and failure load. The response surface methodology was applied, and a new meta-heuristic of optimization called the multi-objective Lichtenberg algorithm was applied to find the optimized configuration of the model. It was possible to increase the failure load by 26.75% in compression performance optimization. Furthermore, in the optimization of modal performance, it was possible to increase the natural frequency by 37.43%. Finally, all 5 responses analyzed simultaneously were optimized. In this case, it was possible to increase the critical buckling load by 42.55%, the failure load by 28.70% and reduce the mass and Poisson's ratio by 15.97 and 11%, respectively. This paper addresses something new in the scientific world to date when evaluating in a multi-objective optimization problem, the compression and modal performance of an auxetic reentrant model.

It was possible to find multi-objective optimized structures. It was possible to increase the critical buckling load by 42.82%, and the failure load in compression performance by 26.75%. Furthermore, in the optimization of modal performance, it was possible to increase the natural frequency by 37.43%, and decrease the mass by 15.97%. Finally, all 5 responses analyzed simultaneously were optimized. In this case, it was possible to increase the critical buckling load by 42.55%, increase the failure load by 28.70% and reduce the mass and Poisson's ratio by 15.97 and 11%, respectively.

There is no work in the literature to date that performed the optimization of 5 responses simultaneously of a reentrant hexagonal cell auxetic structure. This paper also presents an unprecedented statistical analysis in the literature that verifies how the design factors impact each of the responses.