An assembly sequence‐planning system for mechanical parts using neural network

Abstract

PurposeIn this paper, an assembly sequence planning system, based on binary vector representations, is developed. The neural network approach has been employed for analyzing optimum assembly sequence for assembly systems.Design/methodology/approachThe input to the assembly system is the assembly's connection graph that represents parts and relations between these parts. The output to the system is the optimum assembly sequence. In the constitution of assembly's connection graph, a different approach employing contact matrices and Boolean operators has been used. Moreover, the neural network approach is used in the determination of optimum assembly sequence. The inputs to the networks are the collection of assembly sequence data. This data is used to train the network using the back propagation (BP) algorithm.FindingsThe proposed neural network model outperforms the available assembly sequence‐planning model in predicting the optimum assembly sequence for mechanical parts. Due to the parallel structure and fast learning of neural network, this kind of algorithm will be utilized to model another types of assembly systems.Research limitations/implicationsIn the proposed neural approach, the back propagation algorithm is used. Various training algorithms can be employed.Practical implicationsThe simulation results suggest that the neural predictor would be used as a predictor for possible practical applications on modeling assembly sequence planning system.Originality/valueThis paper discusses a new modelling scheme known as artificial neural networks. The neural network approach has been employed for analyzing feasible assembly sequences and optimum assembly sequence for assembly systems.

Local Heuristics Analysis in the Automatic Computation of Assembly Sequences for Building Structures with Multiple Aerial Robots

“…There are other algorithms based on different techniques: ant colonies [21,19,6], neural networks [17], evolutionary algorithms [5,13] and Immune Optimization Approaches(IOA) [3]. Alternatively to perform a graph search, reference [14] proposes to transform the liaisons graph into a table of liaisons in matrix form where a feasible sequence can be found by successively deleting the columns of the parts already included in the assembly and examining their rows for other candidate liaisons to be established.…”

A Sub-assembly Identification Algorithm for Assembly Sequence Planning

“…Through recognizing sub-assemblies of different sizes in a complex assembly, a hierarchical assembling structure of the assembly can be formed. Thus, the complex assembly sequence planning problem is decomposed into several sub-problems, reducing assembly difficulty, facilitating the generation of assembly sequences, and improving assembly planning efficiency [1] . That can better guide product assembling process and reduce assembly costs.…”

Survey on assembly sequencing: a combinatorial and geometrical perspective

“…As usual in NN, energy functions can be defined related to the input and output values of the neurons, and in the present case they are formulated so as to correspond to an optimal assembly sequence when a global optimum of these functions is reached. Other more recent works using back-propagation neural networks for assembly sequence optimization are Cem Sinanoglu (2005) and Chen et al (2008).…”

“…There are academic toy assemblies (a 5-part 2D assembly in Jiménez and Torras (2000) and Ben-Arieh et al (2004), or a 20-part hypothetical product in Motavalli and Islam (1997)), but the majority prefer to show experiments on real assemblies. These range from a 4-part pince in Cem Sinanoglu (2005) to the 48-part gear-box in Chen and Liu (2001). Grouped by type of algorithm (the number of parts shown in parentheses), these examples include:…”

“…Petri nets a flashlight (4) Caselli and Zanichelli (1995) and a ball-point pen (5) Suzuki et al (1993); 2. simulated annealing a relay (10) Hong and Cho (1999); 3. neural networks a pince (4), a hinge (4) and a coupling system (7) Cem Sinanoglu (2005), a gearbox (10) Chen (1992), a relay (10) and an alternator (13) Hong and Cho (1995), and an electric torch (16) Chen et al (2008); 4. genetic algorithms an oil pump (5) Bonneville et al (1995), an industrial controller (19) Guan et al (2002), an hydraulic linear motor (25) Marian et al (2006), an air condition control (28) Sebaaly et al (1996), a signaling relay ( This list -which by no means pretends to be exhaustive-should be interpreted as an illustration of the variety of assemblies dealt with in the existing literature, not as a ranking of the suitability of the different sequencing and optimization algorithms. Furthermore, the examples provided by the authors are generally more oriented towards explaining how their algorithms work than to demonstrate their performance.…”