Open Access. © 2020 E. Kiseleva et al., published by De Gruyter. This work is licensed under the Creative Commons Attribution
4.0 License
Open Comput. Sci. 2020; 10:124–136
Research Article
Elena Kiseleva, Olha Prytomanova, and Liudmyla Hart*
Solving a Two-stage Continuous-discrete Problem
of Optimal Partitioning-Allocation with Subsets
Centers Placement
https://doi.org/10.1515/comp-2020-0142
Received Sep 04, 2019; accepted Apr 06, 2020
Abstract: A two-stage continuous-discrete optimal
partitioning-allocation problem is studied, and a method
and an algorithm for its solving are proposed. This prob-
lem is a generalization of a classical transportation prob-
lem to the case when coordinates of the production points
(collection, storage, processing) of homogeneous products
are continuously allocated in the given domain and the
production volumes at these points are unknown. These
coordinates are found as a solution of the corresponding
continuous optimal set-partitioning problem in a finite-
dimensional Euclidean space with the placement (finding
coordinates) of these subsets’ centers. Also, this problem
generalizes discrete two-stage production-transportation
problems to the case of continuously allocated consumers.
The method and algorithm are illustrated by solving two
model problems.
Keywords: infinite-dimensional mathematical program-
ming, the theory of optimal set-partitioning, transporta-
tion problem, non-differentiable optimization, Shor’s r-
algorithm.
2010 Mathematics Subject Classification: 49M29, 49J52,
49K35
*Corresponding Author: Liudmyla Hart: Department of Computa-
tional Mathematics and Mathematical Cybernetics, Oles Honchar
Dnipro National University (DNU), Dnipro, 49010, Ukraine; Email:
ll_hart@ukr.net
Elena Kiseleva: Faculty of Applied Mathematics, Oles Honchar
Dnipro National University (DNU), Dnipro, 49010, Ukraine; Email:
kiseleva47@ukr.net
Olha Prytomanova: Department of Computational Mathematics
and Mathematical Cybernetics, Oles Honchar Dnipro National
University (DNU), Dnipro, 49010, Ukraine; Email: olgmp@ua.fm
1 Introduction
The mathematical theory of optimal set-partitioning (OSP)
in n-dimensional Euclidean space is a new non-classical
area of infinite-dimensional mathematical programming
operating with Boolean variables.
Studies on the optimal set-partitioning topic were
started in the 1970s by scientists H. Corley and S. Roberts
[1, 2] and, independently, by mathematicians I. Beyko and
E. Kiseleva [3, 4]. In their studies, H. Corley and S. Roberts
were only able to get the necessary conditions for optimal-
ity of partitions. At the same time, I. Beyko and E. Kise-
leva managed to advance further and to develop numer-
ical algorithms for finding optimal solutions based on
similar necessary optimality conditions obtained by them.
This was made possible by using effective methods of
non-differentiable optimization (various variants of the
Shor’s r-algorithm [5–8]) for the numerical solution of aux-
iliary finite-dimensional optimization problems that arise
during the development of methods for solving infinite-
dimensional OSP problems.
Numerous important theoretical and practical opti-
mization problems, that are completely different, are re-
ducible to continuous models of optimal set-partitioning.
Here, let us indicate a few of them: the generalized
Neumann-Pearson problems [9]; various global optimiza-
tion problems [10]; a problem of forming optimal quadra-
tures [9]; a problem of constructing the Dirichlet-Voronoi
diagram and its generalizations [9, 11, 12]; optimal ball cov-
ering continuous problems [13] and other optimal geomet-
ric objects’ packing [14, 15], etc.
Also, the OSP theory can be used for solving applied
problems such as pattern recognition with a goal to min-
imize the average function of wrong recognition; medical
diagnostics aiming to minimize errors of diagnosis; service
sector territorial planning; geological forecasting; environ-
ment protection, e.g., a problem of providing environmen-
tal safety of placing radioactive waste disposal facilities
of nuclear power plants, taking into account an ecologi-
cal structure of the placement domain; placement of am-