Optimal design of fixture layout in a multi-station assembly using highly optimized tolerance inspired heuristic

A Comprehensive Survey of Sensor Deployment Strategies to Diagnose Discrete-Part Manufacturing System

“…For the purpose of deviation diagnosis in a multistation assembly process, optimal sensor allocation methodologies are usually developed base on a state-space model [16,[22][23][24][25][26][27][28][29][30][31][32][33][34][35][36]. As shown in Fig.…”

“…Where μ=[μ1,1(Δ) μ1,1(Δ) ¼μL,nL(Δ) ] T and Cmax is the budget for the total tooling fabrication cost. Tyagi et al [28,31] developed an optimal design of fixture layout to minimize the sensitivity S max while satisfying the geometric and other constraints. The optimization problem can be expressed as follows…”

“…After establishing the optimization basis, a variety of optimization algorithms, from mathematic programming to heuristic search [28,59], have been employed to optimize the sensor deployment.…”

“…Liu et al [48] developed a Best Allocation Subsets by Intelligent Search (BASIS) to obtain the optimal sensor allocation at minimum cost under different specified Average Run Length (ARL) requirements in a manufacturing system. In addition, highly optimized tolerance (HOT) algorithms [28], simulated annealing [30,81], and shuffled frog leaping algorithm [45],ect., are also proposed to deal with the multi-objective sensor optimization distribution problem. Figure 7.…”

A Review of Sensor Layout for Condition Monitoring during Discrete-part Manufacturing

“…The reliability of sensor detection system has different standards according different object, such as diagnosability, observability, uncertainty and so on. Yu [3,30,31] propose a sensor layout method used for the multi-station assemble process. By building the state space model to fuse the sensor data, and then confirming the layout and quantity of sensor with a back propagation algorithm, which use the error propagation among station and error diagnosability in single-station as the indicators of sensor network performance For the partial observed discrete event systems, Jing [32] raise a methods used to select the set of sensor optimizations, Which can provide sufficient information and meet the observability of smallest event.…”