When less is better: Insights from the product mix dilemma from the Theory of Constraints perspective

Abstract

Perhaps due to its origins in a production scheduling software called Optimised Production Technology (OPT), plus the idea of focusing on system constraints, many believe that the Theory of Constraints (TOC) has a vocation for optimal solutions. Those who assess TOC according to this perspective indicate that it guarantees an optimal solution only in certain circumstances. In opposition to this view and founded on a numeric example of a production mix problem, this paper shows, by means of TOC assumptions, why the TOC should not be compared to methods intended to seek optimal or the best solutions, but rather sufficiently good solutions, possible in non-deterministic environments. Moreover, we extend the range of relevant literature on product mix decision by introducing a heuristic based on the uniquely identified work that aims at achieving feasible solutions according to the TOC point of view. The heuristic proposed is tested on 100 production mix problems and the results are compared with the responses obtained with the use of Integer Linear Programming. The results show that the heuristic gives good results on average, but performance falls sharply in some situations.

Production Economics with the Use of Theory of Constraints

“…The POOGI heuristics met with harsh criticism from academic community. The main reason of academic scepticism is the manner of products/jobs selection targeted to the constraint(s) utilization which does not give the optimal solution associated with maximizing net profit and return of investment in the long-term period (Ifandoudas & Gurd, 2010), (Souza, Sobreiro, Nagano & Manfrinato, 2013). In addition, the theory of constraints does not disclose the relationship between the operating activities of the production company and the valuation of products, although the price of each product is one of the key parameters for throughput calculation Ti.…”

“…Data for this example (in particular, unit processing times) are established in such a way that one of the machines is overloaded in a given period of time (there is always at least one machine/constraint incapable of producing all the products P and Q as results from demand). The problem is to decide on the number of products P and Q directed to the production in order not to exceed the total available time and to maximize the profit of the company (Souza et al, 2013). With the development of the theory of constraints PQ issue has evolved.…”

Product Portfolio Decision-making Method Based on Multiple Bottlenecks

“…Then Aryanezhad [6] and others tested RTOC, and found that RTOC sometimes got non-optimal solution, and optimized RTOC. De Souza et al [7] discussed the controversial issue of comparing TOC with the optimization method of product portfolio decision, and put forward a heuristic method to obtain feasible solutions from the perspective of TOC. The results are compared with those of integer linear programming.…”

Decisão de mix de produtos sob a perspectiva do custeio baseado em atividades e tempo para operações com múltiplas restrições.

“…Por exemplo, apesar de não citarem Goldratt (1990a), Hopp e Spearman (2001, p. 548-552) abordam o problema dos produtos P e Q para mostrar diferenças entre os resultados obtidos com a heurística da TOC e os obtidos através da programação linear, chegando à conclusão que a heuríitica da TOC pode resultar em resultados subótimos. Hodiernamente, o problema dos produtos P e Q ainda é utilizado em pesquisas acadêmcias sobre decisão de mix de produtos (ex: SOBREIRO;SOBREIRO;SOUZA et al, 2013), principalmente no que tange ao desenvolvimento de novas heurísticas e procedimentos matemáticos (segunda vertente de pesquisa sobre decisão de mix de produtos). "…”