Better Assembly Through Software

There’s a limit to how much tolerance assembly processes can tolerate. There’s also a limit to how much the people involved in assembly operations can tolerate. Poorly designed parts, assemblies, and, yes, work areas can lead to poor product quality, increased assembly times, and added costs in scrap, rework, and warranties—plus potential personal injuries. Software for dimensional analysis and ergonometric analysis should be in every manufacturer’s “toolkit.” Both analysis tools go beyond incrementally improving assembly operations; they directly improve a company’s profitability. 

Focus on assembly variation
Software for dimensional analysis ensures parts fit and work together properly in an assembly. The software predicts what problems might exist when building the assembly—before actually making and assembling the physical parts. The analysis, specifically tolerance stack-ups (the cumulative effect of individual part tolerances in an assembly), is critical to addressing mechanical fit and performance, as well as catching design flaws before investing in expensive tooling. Mananufacturing costs are also reduced by maximizing part tolerances that can be increased, while controlling the overall dimensional integrity of an assembly.

One such analysis tool is Tecnomatix Variation Analysis from Siemens PLM Software (siemens.com/plm). The software uses the visualization capabilities in the Teamcenter PLM system to simulate the process of assembling parts into subassemblies and finished products. The simulation includes a 3D digital prototype of part geometry, product variations (tolerances), assembly process variations (sequence, assembly attachment definition, tooling), and measurements. The simulation verifies tolerances and dimensioning schemes, often identifying critical dimensions, tolerances, and assembly processes that significantly contribute to variation. Different options and iterations can be simulated to optimize part designs, tolerances, and the assembly process itself, or all three. The software can also determine if an assembly process is constrained too much or too little. 

While the Siemens’s product is CAD-neutral, there are also variation analysis tools from non-CAD vendors. For example, Sigmetrix (sigmetrix.com) sells CETOL 6σ, which works from the annotations in the solid models in the CAD systems from Dassault Systèmes Catia, PTC Creo and Pro/Engineer, and SolidWorks. CETOL is fully associative with those systems; changes made to the CAD annotations are propagated to CETOL and, conversely, changes made within CETOL in the process of a tolerance analysis are written back to the CAD models.

Root-sum-square (RSS) is the statistical method of choice for analyzing toler-ces for many companies. RSS, according to Sigmetrix, sums up dimension distributions, not the tolerances themselves; worst-case analysis is on process distribution moments (e.g., standard deviation) rather than design tolerances. According to the company, “Statistical analysis (also called variation analysis) can be used to predict the actual variation of an assembly based on the variation of the part dimensions. This approach requires distributions to be normal with all parts at the same quality level, i.e. ±3s.” CETOL, however, uses “the method of system moments.” This method can analyze all complexities of a design with no restriction on distribution type or quality level. As a result, users can perform full assembly variation analysis with the tolerance analysis software.

Focus on assembly personnel
The people who put an assembly together are another critical part of any assembly operation. Software for workplace ergonomic analysis simulates virtualized people working in a virtualized production area. Ergonomic specialists can experiment with different options, redesign and relocate production machinery and related equipment, redefine work tasks, and change other aspects of the assembly environment across a range of human factors. This analysis helps companies analyze and predict human safety and performance, then take the necessary steps to reduce the potential for worker fatigue, injuries and accidents.

For instance, Delmia ergonomics software from Dassault Systèmes (3ds.com)—including Ergonomics Task Definition (ETD), Ergonomics Evaluation (EGE), and Ergonomics Analysis (EGA)—lets ergonomic specialists put lifelike, digital human models in 3D models of the workplace. The software lets specialists examine, compare, and rate progressively different postures and movements of people (whole body and individual limbs, such as arms, wrists, and legs), as well as validate whether individual work tasks are even possible.

ETD includes predefined worker actions, such as picking up and placing objects, walking, operating a device, and using specific tools. The anthropometric definition of the manikins can be “standard” or based on specific populations of people. For example, EGE lets specialists create manikins based on the 95th-percentile American male, but it can predict whether a 95th-percentile Japanese male would be able to perform the same task. Supported populations include Canada, France, Japan, Korea, and the United States. (Customized population databases and manikin anthropometry are also possible.) Among the functions of EGA is the detection of work-related upper-limb disorder risks through the assessment of the biomechanical and postural workloads on a person, focusing specifically on the neck, trunk, and upper limbs. Color-coded analysis results are displayed on the manikin’s upper body segments (neck, trunk, wrists, and arms); analysis reports can be exported in text or HTML. EGA can also analyze lifting and lowering tasks based on variables such as task duration and frequency, lifting posture (start and finish), and coupling conditions. The analysis provides recommended and maximum weights for lifted objects. Based on the manikin’s position and the specified load (the object’s weight) on the manikin’s limbs, the biomechanical analysis will calculate the moments and forces applied to each human joint. This analysis will also determine the percentage of the manikin’s population that will be unable to perform the task in question.