Revista de la Facultad de Ingeniería U.C.V., Vol. 32, N°10, pp. 443-449, 2017
443
Exploration of An Optimized Comprehensive Algorithm for Goods
Stacking Based on The Simulated Annealing Algorithm and Genetic
Algorithm
Peng Li*, Yong Tang, Bin Yang
College of Automobile and Traffic Engineering, Nanjing Forestry University, Nanjing, Jiangsu, 210037,
;
China
*Corresponding author (E-mail: lipengaq@nuaa.edu.cn)
Abstract
Along with intensification of market competition, and in order to guarantee steady and efficient operation of
production and to obtain the maximum economic interests, the original simple, partial and conventional control
and experience-based management can no longer meet requirements of modern production. All corporate
managers and control engineers are faced with the following problems, including how operation decision-
making and organization of production should change along with changes of the raw material supply and
product demands on the market; how the production process should be controlled under the condition that the
production plan is changed so that production flexibility can be maximized; how management and decision-
making should be carried out under the prerequisite of not dramatically changing the production process so that
comprehensive economic interests of corporate production can be maximized. In the highly automatic system,
reasonable and efficient operation of the production process has become extremely complex. An effective set of
computer scheduling and control strategies is required. This justifies the necessity of studying the issue of
scheduling. Since most scheduling issues are non-deterministic polynomial complete problems (NP), there have
not yet been any efficient solution strategies. Thus, research into the issue of scheduling also holds huge
theoretical significance. This paper finds out that the simulated annealing algorithm (SAA) can give full play to
advantages of the simulation algorithm and the genetic algorithm. This can not only narrow down the search
range of feasible solutions, but also avoid getting stuck in the partial optimal solution. During the practical
operation process, the SAA can effectively improve the space utilization rate of the tank, and provide vigorous
theoretical basis for reduction of corporate costs and human resource waste.
Key words: Annealing Algorithm, Genetic Algorithm, Optimized Comprehensive Algorithm
1. INTRODUCTION
Along with development of science and technology, the production scale has been expanding and
becoming increasingly complex(Hui,2010). The market competition has been intensifying(Ghezavati and
Nia,2015). All this has raised a higher requirement of corporate management and monitoring of the production
process(Jonathan, Bruno and Hamid, 2010). In the recent decades, various production processes have undergone
dramatic changes(Nallakumarasamy, Srinivasan and Raja, 2011) . Large-scale production and continuity of the
production process have become two striking characteristics (Liu and Ye, 2014). In order to solve the above
problems, Dr. Harrington proposed the concept of computer integrated manufacturing in 1973(Liu, Sun and
Yan , 2011). Computer integrated manufacturing or CIM is a new philosophy to organize, manage and operate
corporate production(Lin, Chang and Lie, 2010). It relies on computer software and hardware, and makes a
comprehensive utilization of modern management techniques, manufacturing techniques, information
techniques, automation techniques, and systematic engineering techniques to organically integrate and optimize
operation of elements related to humans, technology and operation as well as information flow and material
flow(Abdi, Fathian and Safari,2012). All in all, CIM can help improve product quality and reduce consumption,
and help enterprises win the market competition. According to the above definition of CIM, CIMS or computer
integrated manufacturing system is a practical system integrated based on philosophy.
A method to obtain the maximum of the utility function and the profit function,
, is to adopt the
objective function as the fitness function. However, many optimization issues aim at obtaining the minimum of
the cost function,
; while the genetic algorithm requires the fitness function to be positive. Meanwhile, the
higher the degree of fitness is, the better the individual will be. Therefore, under many occasions, the objective
function of problems is adopted as the measure of the objective function. Apart from transforming the objective
function into the form of maximum, the fitness function should be non-negative. Below is a major method for
conversion:
(1)