W pracy zaproponowany został model planowania opłacalnego ekonomicznie demontażu. Model ten należy do ścisłych metod generowania sekwencji operacji, wykorzystuje bowiem ścisłe algorytmy wyznaczania sekwencji demontażowej, a wybór sekwencji optymalnej dokonywany jest za pomocą programowania dynamicznego. Proces demontażu podzielono na dwa etapy: 1) wymuszony, w którym wydziela się część wartościową lub szkodliwą dla otoczenia bądź część wymienianą w ramach naprawy całego wyrobu; 2) możliwy, będący kontynuacją demontażu wymuszonego, którego celem jest uzyskanie dalszych ekonomicznych korzyści. Znalezienie optymalnej sekwencji demontażowej nie jest jedynym atutem przedstawionego systemu projektowania procesu. Dzięki temu systemowi łatwe staje się rozważanie alternatywnych struktur projektowanego wyrobu. Analizy dokonane za jego pomocą umożliwiają sformułowanie zaleceń dla modyfikacji już istniejących projektów wyrobów.
In this paper, we have proposed a model of planning a profitable disassembly process. The model belongs to the category of strict operation sequence generation methods since it uses strict algorithms of designating disassembly sequences, while the choice of an optimum sequence is made with the use of dynamic programming. The disassembly process is subdivided into two stages: 1) enforced process to identify either valuable, or hazardous, or replaceable parts in case of the product repair; 2) optional process, being a continuation of enforced disassembly, whose goal is to obtain further economic benefits. The prohibition of direct dumping of a whole product in a landfill causes the necessity to identify a set of all possible disassembly sequences. The whole concept is based on the enumeration of all the unit decompositions, or determination of a set of cuts to be made in the graph that models geometric, topological and material structures of the product, followed by the selection of those cuts which fulfill the operation feasibility conditions. The set of technically possible disassembly operations has been extended by the operations that destroy particular parts of the unit being disassembled, disintegrate parts and subassemblies, and separate preselected materials. The set of disassembly operations has been recorded using a digraph of <Qx, Lx> in which vertices Qx correspond to the states of divisions of the product being disassembled, while arcs Lx model disassembly operations. The tool used for resolving the disassembly planning problem was dynamic programming that finds an optimum way in the directed graph of disassembly sequences. The model presented here allows to select the optimum solution with respect to disassembly cost, course or degree. The procedure determines the moment of process interruption at which further benefits of the disassembly process may not be expected. Optimization of decisions at each possible state of disassembly should be treated as a parameter problem. That is the reason of the exponential convergence of the computation process and, by the same, of the limitation of the size of problems being resolved to the units composed of 20-30 parts at most. Finding an optimum disassembly sequence is not the only advantage of the process design system presented here. Owing to our system, it is easy to consider alternative structures of the product being designed. The analyses conducted with the model application made it possible to formulate suggested recommendations for the modification of the existing product designs. They include the identification of elements (parts and materials) connected in groups that maintain material uniformity, or those that must be removed to a landfill.