Processing math: 100%
Research article Special Issues

Optimal covering points and curves

  • We consider the optimal covering of the unit square by N circles. By optimal, we mean the covering that can be done with N circles of minimum radius. Equivalently, we study the problem of the optimal placement of N points such that the maximum over all locations in the square of the distance of the location to the set of points is minimized. We propose a new algorithm that can identify optimal coverings to high precision. Numerical predictions of optimal coverings for N = 1 to 16 agree with the best known results in the literature. We use the optimal designs in approximations to two novel, related problems involving the optimal placement of curves.

    Citation: Justin Tzou, Brian Wetton. Optimal covering points and curves[J]. AIMS Mathematics, 2019, 4(6): 1796-1804. doi: 10.3934/math.2019.6.1796

    Related Papers:

    [1] Yaser Saber, Hanan Alohali, Tawfik Elmasry, Florentin Smarandache . On single-valued neutrosophic soft uniform spaces. AIMS Mathematics, 2024, 9(1): 412-439. doi: 10.3934/math.2024023
    [2] Sumyyah Al-Hijjawi, Abd Ghafur Ahmad, Shawkat Alkhazaleh . A generalized effective neurosophic soft set and its applications. AIMS Mathematics, 2023, 18(12): 29628-29666. doi: 10.3934/math.20231517
    [3] Majdoleen Abu Qamar, Abd Ghafur Ahmad, Nasruddin Hassan . An approach to Q-neutrosophic soft rings. AIMS Mathematics, 2019, 4(4): 1291-1306. doi: 10.3934/math.2019.4.1291
    [4] Muhammad Ihsan, Muhammad Saeed, Atiqe Ur Rahman, Mazin Abed Mohammed, Karrar Hameed Abdulkaree, Abed Saif Alghawli, Mohammed AA Al-qaness . An innovative decision-making framework for supplier selection based on a hybrid interval-valued neutrosophic soft expert set. AIMS Mathematics, 2023, 8(9): 22127-22161. doi: 10.3934/math.20231128
    [5] Khulud Fahad Bin Muhaya, Kholood Mohammad Alsager . Extending neutrosophic set theory: Cubic bipolar neutrosophic soft sets for decision making. AIMS Mathematics, 2024, 9(10): 27739-27769. doi: 10.3934/math.20241347
    [6] Hatice Tasbozan . Near neutrosophic soft set. AIMS Mathematics, 2024, 9(4): 9447-9454. doi: 10.3934/math.2024461
    [7] Rana Muhammad Zulqarnain, Wen Xiu Ma, Imran Siddique, Shahid Hussain Gurmani, Fahd Jarad, Muhammad Irfan Ahamad . Extension of aggregation operators to site selection for solid waste management under neutrosophic hypersoft set. AIMS Mathematics, 2023, 8(2): 4168-4201. doi: 10.3934/math.2023208
    [8] Tareq M. Al-shami, Abdelwaheb Mhemdi, Alaa M. Abd El-latif, Fuad A. Abu Shaheen . Finite soft-open sets: characterizations, operators and continuity. AIMS Mathematics, 2024, 9(4): 10363-10385. doi: 10.3934/math.2024507
    [9] Alaa M. Abd El-latif . Novel types of supra soft operators via supra soft sd-sets and applications. AIMS Mathematics, 2024, 9(3): 6586-6602. doi: 10.3934/math.2024321
    [10] Orhan Göçür . Amply soft set and its topologies: AS and PAS topologies. AIMS Mathematics, 2021, 6(4): 3121-3141. doi: 10.3934/math.2021189
  • We consider the optimal covering of the unit square by N circles. By optimal, we mean the covering that can be done with N circles of minimum radius. Equivalently, we study the problem of the optimal placement of N points such that the maximum over all locations in the square of the distance of the location to the set of points is minimized. We propose a new algorithm that can identify optimal coverings to high precision. Numerical predictions of optimal coverings for N = 1 to 16 agree with the best known results in the literature. We use the optimal designs in approximations to two novel, related problems involving the optimal placement of curves.



    Data is a valuable source of knowledge that contains helpful information if exploited effectively [1]. One of the challenges facing data researchers is the ambiguity and uncertainty of the data they have access to, which makes it difficult for them to process information. But these challenges are, in a positive sense, opportunities for the development of new techniques and tools, such as they various approaches based on fuzzy set theory [2]. The advent of fuzzy theory has prompted extensive work on ideas such as fuzzy sets [3], vague sets [4], soft sets [5], and neutrosophic sets [6]. It was originally thought that the development of new theories would eclipse fuzzy theory, but that does not seem to be the case [7]. This research field is becoming more and more active, with a number of fundamental contributions to the rapid development of new theories [8,9]. One of the most prominent applications is the use of fuzzy set theory in emerging and vibrant fields like machine learning [10,11] or topological data analysis [12,13].

    In recent years, the study of soft sets [5] and neutrosophic sets [14] has become an attractive research area. Neutrosophic sets recently emerged as a tool for dealing with imprecise, indeterminate, and inconsistent data [15]. In contrast, soft sets show potential for dealing with uncertainties that classical methods cannot control [16]. Combining these two types of sets results in a unique hybrid structure, a neutrosophic soft set (NS-set) [17], for working effectively in uncertain environments. Maji proposed this [17,18] in 2013 and it was modified by Deli and Broumi [19] in 2015. Furthermore, Karaaslan [20] redefined this concept and its operations to be more efficient and complete. Since then, this structure has proved to be quite effective when applied in real life in many fields, such as decision making [17], market prediction [21], and medical diagnosis [22,23].

    The topology on NS-sets is one of the issues that needs more attention, alongside neutrosophic topology [24,25] and soft topology [26]. This issue has emerged recently to help complete the overall picture for NS and aid its practical applications based on topology [27,28]. In 2017, Bera and Mahapatra [29] gave general operations to construct a topology on NS-sets. They also presented concepts related to topological space such as interior, closure, neighborhood, boundary, regularity, base, subspace, separation axioms, along with specific illustrations and proofs. In 2018, these authors [30] continued to develop further studies on connectedness and compactness on NS-topological space. In 2019, Ozturk, Aras, and Bayramov [31] introduced a new approach to topology on NS-sets. This approach is quite different from the previous work [29], and was further developed by constructing separation axioms [32] in the same year, 2018. Recently, the continuum [33] or compactness [34] on the topological space generated on NS-sets has also been studied with the same properties as the normal space. Many variations [35] of the topological space on NS-sets have also attracted the attention of researchers, and most of the related works are inspired by topology on neutrosophic and soft sets with the idea of a hybrid structure [36,37].

    In this work, we construct the topological space and related concepts on NS-sets through general operations in a way that is very different from the work of Bera and Mahapatra [29,30], but more general than the work of Ozturk, Aras, and Bayramov [31,32], with our operations based on the generality of min and max operations. This work begins by defining two new operations to create the relationships between NS-sets. These relations are then used as the kernel for forming topology and topological relations on NS-sets. One emphasis shown here is on elucidating the relationship between the topology on NS-sets and the component fuzzy soft topologies. All the ideas in this work are presented convincingly and clearly through definitions, theorems, and their consequences.

    In summary, the significant contributions of this study are as follows:

    (1) Defining two novel concepts, called minnorm and maxnorm, to provide a theoretical foundation for determining operations on NS-sets, including intersection, union, difference, AND, and OR.

    (2) Constructing the topology, open set, closed set, interior, closure, and regularity concepts on NS-sets based on just determined operations.

    (3) Elucidating the relationship between the topology on NS-sets and the fuzzy soft topologies generated by truth, indeterminacy, falsity degrees by the theorems and counterexamples.

    (4) The concepts are well-defined, and the theorems are proved convincingly and logically.

    This work is organized as follows: Section 1 presents the motivation and introduces the significant contributions. Section 2 briefly introduces NS-sets and related concepts. The two new ideas, minnorm and maxnorm, are provided in Section 3 as a theoretical foundation for determining operations on NS-sets, including intersection, union, difference, AND, and OR. In Section 4, the topology on NS-sets is defined with related concepts such as open set, closed set, interior, closure, and regularity. Furthermore, the relationship between the topology on NS-sets and the fuzzy soft topologies generated by truth, indeterminacy, and falsity functions by theorems and counterexamples in Section 5. The last section presents conclusions and future research trends in this area.

    This section recalls the NS-set proposed in 2013 by Maji [17,18], then modified and improved in 2015 by Deli and Broumi [19]. This concept is based on combining soft [5] and neutrosophic [6] sets. Some background related to NS-sets is briefly presented below so that readers can better understand the following sections.

    Without loss of generality, we consider X to be a universal set, E to be a parameter set, and N(X) to denote the collection of all neutrosophic sets on X.

    Definition 1. ([18,19]). The pair (A,E) is a NS-set on X where A:EN(X) is a set valued function determined by eA(e)Ae with

    Ae:X]0;1+[×]0;1+[×]0;1+[
    xAe(x)TAe(x);IAe(x);FAe(x) (1)

    for all eE, and the real function triples TAe,IAe,FAe:X]0;1+[ indicate truth, indeterminacy, and falsity degrees, respectively, with no restriction on their sum.

    In other words, the NS-set can be described as a set of ordered tuples as follows:

    (A,E)={(e,A(e)):eE,A(e)N(X)} (2)
    ={(e,x,TAe(x),IAe(x),FAe(x)):eE,xX} (3)
    :={(e,xTAe(x),IAe(x),FAe(x)):eE,xX}. (4)

    If nothing changes, the symbol NS(X) indicates the collection of all NS-sets on X. Besides, if the NS-sets consider the same parameter set E, then it is not mentioned repeatedly in order to simplify the notations. Moreover, because the values of T, I, F belong to the unit interval [0;1], the integral part of the values is almost zero. Typically, it may occur that the integer part is omitted (for example, .1 instead of 0.1). Therefore, if it does not lead to confusion, this omitted format of a decimal is always used in all the tables used in this paper.

    Definition 2. ([18,19]).

    a. E is a null NS-set if

    eE,xX,{TE(x)=0IE(x)=0FE(x)=1. (5)

    b. ˜E is a semi-null NS-set if

    eE,xX,{T˜E(x)=0I˜E(x)=0F˜E(x)=1. (6)

    c. XE is an absolute NS-set if

    eE,xX,{TXE(x)=1IXE(x)=1FXE(x)=0. (7)

    d. X˜E is a semi-absolute NS-set if

    eE,xX,{TX˜E(x)=1IX˜E(x)=1FX˜E(x)=0. (8)

    Definition 3. ([19,31]). Let A and B be two NS-sets on X.

    a. A is a NS-subset of B,writtenasAB, if

    eE,xX,{TAe(x)TBe(x)IAe(x)IBe(x)FAe(x)FBe(x). (9)

    b. A is a NS-superset of B,writtenasAB, if B is a NS-subset of A.

    c. ¯A is the complement of A if

    eE,xX,{T¯Ae(x)=FAe(x)I¯Ae(x)=1IAe(x)F¯Ae(x)=TAe(x). (10)

    Example 1. Let two NS-sets M and N be represented in Table 1 as follows:

    Table 1.  NS-sets M and N.
    M e1 e2 N e1 e2
    x1 .2,.3,.4 .3,.5,.5 x1 .3,.6,.1 .4,.5,.4
    x2 .3,.4,.3 .6,.2,.4 x2 .6,.5,.2 .7,.3,.2
    x3 .3,.5,.2 .4,.4,.3 x3 .4,.5,.3 .6,.3,.3
    x4 .2,.7,.6 .3,.4,.3 x4 .9,1,.4 .4,.5,.1

     | Show Table
    DownLoad: CSV

    Based on Eq (9) of Definition 3, MN.

    Example 2. The NS-set P and its complement ¯P are represented according to Eq (10) in Table 2 as follows:

    Table 2.  NS-sets P and ¯P.
    P e1 e2 e3 ¯P e1 e2 e3
    x1 .9,.8,.2 .3,.7,.2 .4,.6,.3 x1 .2,.2,.9 .2,.3,.3 .3,.4,.4
    x2 .7,.6,.2 .3,.4,.6 .4,.3,.2 x2 .2,.4,.7 .6,.6,.3 .2,.7,.4
    x3 .3,.3,.5 .1,.2,.3 .9,.5,.8 x3 .5,.7,.3 .3,.8,.1 .8,.5,.9

     | Show Table
    DownLoad: CSV

    Theorem 1. If ANS(X),

    (1) ¯¯A=A,

    (2) ¯E=XE,

    (3) ¯˜E=X˜E,

    (4) ¯XE=E,

    (5) ¯X˜E=˜E.

    Proof. These properties are directly inferred from the definitions of the null, semi-null, absolute, semi-absolute NS-sets and the complement operation.

    In this section, we focus on defining two novel norms, called minnorm and maxnorm, as the foundations for determining operations on NS-sets in general. Each operation is well-defined along with its well-proven properties.

    Definition 4. A minnorm is the binary operation :[0;1]×[0;1][0,1] that obeys the conditions as follows:

    (a) has the commutative and associative properties,

    (b) x[0,1],x1=1x=x,

    (c) x[0,1],x0=0x=0,

    (d) x,y[0,1],xxy.

    Definition 5. A maxnorm is the binary operation :[0;1]×[0;1][0,1] that obeys the following conditions:

    (a) has the commutative and associative properties,

    (b) x[0,1],x1=1x=1,

    (c) x[0,1],y0=0x=x,

    (d) x,y[0,1],xxy.

    Definition 6. The minnorm and maxnorm satisfy De Morgan's law if they obey the following conditions:

    x,y[0,1],(1x)(1y)=1xy, (11)
    x,y[0,1],(1x)(1y)=1xy. (12)

    Some commonly used minnorm and maxnorm are shown in Table 3. On the other hand, all of these norms satisfy De Morgan's law in pairs.

    Table 3.  Some commonly used minnorm and maxnorm satisfying the De Morgan's law.
    minnorms maxnorms
    1 x,y[0,1],xy=xy x,y[0,1],xy=x+yxy
    2 x,y[0,1],xy=min{x,y} x,y[0,1],xy=max{x,y}
    3 x,y[0,1],xy=max{x+y1,0} x,y[0,1],xy=min{x+y,1}

     | Show Table
    DownLoad: CSV

    Definition 7. The intersection of the two NS-sets A and B,writtenasAB, is determined by

    eE,xX,{TABe(x)=TAe(x)TBe(x)IABe(x)=IAe(x)IBe(x)FABe(x)=FAe(x)FBe(x). (13)

    Example 3. Let two NS-sets A and B be represented in Table 4 as follows:

    Table 4.  NS-sets A and B.
    A e1 e2 B e1 e2
    x1 .2,.4,.5 .5,.2,.8 x1 .7,.2,.7 .6,0,.5
    x2 .1,.4,.3 .8,.9,.4 x2 .3,.9,.1 .7,.7,.9
    x3 .1,.2,.7 .8,.9,.4 x3 .6,.4,.7 .8,.1,0

     | Show Table
    DownLoad: CSV

    If using minnorms xy=max{x+y1,0} and maxnorms xy=min{x+y,1}, the intersection AB of the two above NS-sets is described according to Eq (13) in Table 5 as follows:

    Table 5.  NS-sets AB.
    AB e1 e2
    x1 0,0,1 .1,0,1
    x2 0,.3,.4 .5,.6,1
    x3 0,0,1 .6,0,.4

     | Show Table
    DownLoad: CSV

    Theorem 2. If A,B,CNS(X),

    (1) AA=A,

    (2) AE=E,

    (3) A˜E=˜E,

    (4) AXE=A,

    (5) AX˜E=A,

    (6) A(BC)=(AB)C,

    (7) AB=BA.

    Proof. These properties are directly inferred from the definitions of norms and intersection operation.

    Definition 8. Let (Ai)iI be a collection of NS-sets on X. The intersection of the collection of NS-sets (Ai)iI,writtenasiIAi, is determined by

    eE,xX,{TiIAie(x)=iI{TAie(x)}IiIAie(x)=iI{IAie(x)}FiIAie(x)=iI{FAie(x)}. (14)

    Definition 9. The union of the two NS-sets A and B,writtenasAB, is determined by

    eE,xX,{TABe(x)=TAe(x)TBe(x)IABe(x)=IAe(x)IBe(x)FABe(x)=FAe(x)FBe(x). (15)

    Example 4. If using minnorms xy=max{x+y1,0} and maxnorms xy=min{x+y,1}, the union AB of the two above NS-sets A and B in Example 3 is described according to Eq (14) in Table 6 as follows:

    Table 6.  NS-sets AB.
    AB e1 e2
    x1 .9,.6,.2 1,.2,.3
    x2 .4,1,1 1,1,.3
    x3 .7,.6,.4 1,1,0

     | Show Table
    DownLoad: CSV

    Theorem 3. If A,B,CNS(X),

    (1) AA=A,

    (2) AE=A,

    (3) A˜E=A,

    (4) AXE=XE,

    (5) AX˜E=X˜E,

    (6) A(BC)=(AB)C,

    (7) AB=BA.

    Proof. These properties are directly inferred from the definitions of norms and union operation.

    Theorem 4. If the minnorm and maxnorm satisfy De Morgan's law, for all A,BNS(X),

    (1) ¯AB=¯A¯B,

    (2) ¯AB=¯A¯B.

    Proof.

    (1) eE,xX,

    {T¯ABe(x)=FABe=FAe(x)FBe(x)I¯ABe(x)=1IABe=1IAe(x)IBe(x)F¯ABe(x)=TABe=TAe(x)TBe(x), (16)

    and

    {T¯A¯Be(x)=T¯Ae(x)T¯Be(x)=FAe(x)FBe(x)I¯A¯Be(x)=I¯Ae(x)I¯Be(x)=(1IAe(x))(1IBe(x))F¯A¯Be(x)=F¯Ae(x)F¯Be(x)=TAe(x)TBe(x). (17)

    Moreover,

    (1IAe(x))(1IBe(x))=1IAe(x)IBe(x), (18)

    due to De Morgan's law of the minnorm and maxnorm. Therefore,

    eE,xX,{T¯ABe(x)=T¯A¯Be(x)I¯ABe(x)=I¯A¯Be(x)F¯ABe(x)=F¯A¯Be(x). (19)

    (2) eE,xX,

    {T¯ABe(x)=FABe=FAe(x)FBe(x)I¯ABe(x)=1IABe=1IAe(x)IBe(x)F¯ABe(x)=TABe=TAe(x)TBe(x), (20)

    and

    {T¯A¯Be(x)=T¯Ae(x)T¯Be(x)=FAe(x)FBe(x)I¯A¯Be(x)=I¯Ae(x)I¯Be(x)=(1IAe(x))(1IBe(x))F¯A¯Be(x)=F¯Ae(x)F¯Be(x)=TAe(x)TBe(x). (21)

    Moreover,

    (1IAe(x))(1IBe(x))=1IAe(x)IBe(x), (22)

    due to De Morgan's law of the minnorm and maxnorm. Therefore,

    eE,xX,{T¯ABe(x)=T¯A¯Be(x)I¯ABe(x)=I¯A¯Be(x)F¯ABe(x)=F¯A¯Be(x). (23)

    The distributive properties between intersection and union operations are not satisfied in the case of these general operations. Counterexamples are shown in Example 5.

    Example 5. Let the NS-set C be represented in Table 7 as follows:

    Table 7.  NS-sets C.
    C e1 e2
    x1 .2,.1,.9 .3,.2,.6
    x2 .3,.7,.6 .8,.2,.5
    x3 .2,.1,.4 .3,.2,.5

     | Show Table
    DownLoad: CSV

    If using minnorm xy=max{x+y1,0} and maxnorm xy=min{x+y,1} with the two above NS-sets A and B in Example 3, the two NS-sets A(BC) and (AB)(AC) can be described in Table 8 as follows:

    Table 8.  NS-sets A(BC) and (AB)(AC).
    A(BC) e1 e2 (AB)(AC) e1 e2
    x1 .1,0,1 .4,0,.9 x1 0,0,1 .1,0,1
    x2 0,1,.3 .8,.8,.8 x2 0,.4,.3 .6,.1,.9
    x3 0,0,.8 .8,.2,.4 x3 0,0,1 .1,.1,.9

     | Show Table
    DownLoad: CSV

    Therefore, A(BC)(AB)(AC). Similarly, see Table 9:

    Table 9.  NS-sets A(BC)(AB)(AC).
    A(BC) e1 e2 (AB)(AC) e1 e2
    x1 .2,.4,.5 .5,.2,.8 x1 .3,.1,.6 .8,0,.7
    x2 .1,1,0 1,.9,.4 x2 0,1,1 1,1,.3
    x3 .1,.2,.7 .9,.9,0 x3 0,0,.5 1,1,0

     | Show Table
    DownLoad: CSV

    Therefore, A(BC)(AB)(AC).

    Definition 10. Let (Ai)iI be a collection of NS-sets on X. The union of the collection of NS-sets (Ai)iI,writtenasiIAi, is determined by

    eE,xX,{TiIAie(x)=iI{TAie(x)}IiIAie(x)=iI{IAie(x)}FiIAie(x)=iI{FAie(x)}. (24)

    Definition 11. The difference of the two NS-sets A and B,writtenasAB, is determined by AB=A¯B, i.e.,

    eE,xX,{TABe(x)=TAe(x)FBe(x)IABe(x)=IAe(x)(1IBe(x))FABe(x)=FAe(x)TBe(x). (25)

    Example 6. If using minnorm xy=max{x+y1,0} and maxnorm xy=min{x+y,1}, the difference AB of the two above NS-sets A and B in Example 3 is described according to Eq (25), see Table 10:

    Table 10.  NS-sets AB.
    AB e1 e2
    x1 0,.2,1 0,.2,1
    x2 0,0,.6 .7,.2,1
    x3 0,0,1 0,.8,1

     | Show Table
    DownLoad: CSV

    Theorem 5. If the minnorm and maxnorm satisfy De Morgan's law, for all A,B,CNS(X),

    (1) ABA,

    (2) ¯AB=¯AB,

    (3) ¯A¯B=BA,

    (4) A(BC)=(AB)(AC),

    (5) (AB)C=(AC)(BC),

    (6) (AB)(CD)=(CB)(AD)=(AC)(BD).

    Proof.

    (1) eE,xX,{TABe(x)=TAe(x)TBe(x)TAe(x)IABe(x)=IAe(x)(1IBe(x))IAe(x)FABe(x)=FAe(x)FBe(x)FAe(x). This implies that ABA.

    (2) ¯AB=¯A¯B=¯A¯¯B=¯AB due to Theorem 1.

    (3) ¯A¯B=¯A¯¯B=¯AB=B¯A=BA.

    (4) A(BC)=A¯BC=A(¯B¯C)=(A¯B)(A¯C)=(AB)(AC) due to Theorems 3 and 4.

    (5) (AB)C=(AB)¯C=(A¯C)(B¯C)=(AC)(BC) due to Theorem 3.

    (6) (AB)(CD)=(A¯B)(C¯D)=(C¯B)(A¯D)=(CB)(AD) due to Theorem 3.

    (7) (AB)(CD)=(A¯B)(C¯D)=(AC)(¯B¯D)=(AC)¯BD =(AC)(BD) due to Theorems 3 and 4.

    Definition 12. The AND operation of the two NS-sets A and B with the same parameter set E, written as AB, is determined over the same parameter set E×E by

    (e1,e2)E×E,xX,{TAB(e1,e2)(x)=TAe1(x)TBe2(x)IAB(e1,e2)(x)=IAe1(x)IBe2(x)FAB(e1,e2)(x)=FAe1(x)FBe2(x). (26)

    Definition 13. The OR operation of the two NS-sets A and B with the same parameter set E, written as AB, is determined over the same parameter set E×E by

    (e1,e2)E×E,xX,{TAB(e1,e2)(x)=TAe1(x)TBe2(x)IAB(e1,e2)(x)=IAe1(x)IBe2(x)FAB(e1,e2)(x)=FAe1(x)FBe2(x). (27)

    Example 7. If using minnorm xy=max{x+y1,0} and maxnorm xy=min{x+y,1}, the AND AB and OR AB operations of the two above NS-sets A and B in Example 3 is described according to Eqs (26) and (27) in Table 11 as follows:

    Table 11.  NS-sets AB and AB.
    AB (e1,e1) (e1,e2) (e2,e1) (e2,e2) AB (e1,e1) (e1,e2) (e2,e1) (e2,e2)
    x1 0,0,1 0,0,1 .2,0,1 .1,0,1 x1 .9,.6,.2 .8,.4,1 1,.4,.5 1,.2,.3
    x2 0,.3,.4 0,.1,1 .1,.8,.5 .5,.6,1 x2 .4,1,1 .8,1,.2 1,1,0 1,1,.3
    x3 0,0,1 0,0,.7 .4,.3,.1 .6,0,.4 x3 .7,.6,.4 .9,.3,0 1,1,.1 1,1,0

     | Show Table
    DownLoad: CSV

    Theorem 6. If the minnorm and maxnorm satisfy De Morgan's law, for all A,BNS(X),

    (1) ¯AB=¯A¯B,

    (2) ¯AB=¯A¯B.

    Proof.

    (1) (e1,e2)E×E,xX,

    {T¯AB(e1,e2)(x)=FAB(e1,e2)=FAe1(x)FBe2(x)I¯AB(e1,e2)(x)=1IAB(e1,e2)=1IAe1(x)IBe2(x)F¯AB(e1,e2)(x)=TAB(e1,e2)=TAe1(x)TBe2(x), (28)

    and

    {T¯A¯B(e1,e2)(x)=T¯Ae1(x)T¯Be2(x)=FAe1(x)FBe2(x)I¯A¯B(e1,e2)(x)=I¯Ae1(x)I¯Be2(x)=(1IAe1(x))(1IBe2(x))F¯A¯B(e1,e2)(x)=F¯Ae1(x)F¯Be2(x)=TAe1(x)TBe2(x). (29)

    Moreover,

    (1IAe1(x))(1IBe2(x))=1IAe(x)IBe(x), (30)

    due to De Morgan's law of the minnorm and maxnorm. Therefore,

    (e1,e2)E×E,xX,{T¯AB(e1,e2)(x)=T¯A¯B(e1,e2)(x)I¯AB(e1,e2)(x)=I¯A¯B(e1,e2)(x)F¯AB(e1,e2)(x)=F¯A¯B(e1,e2)(x). (31)

    (2) (e1,e2)E×E,xX,

    {T¯AB(e1,e2)(x)=FAB(e1,e2)=FAe1(x)FBe2(x)I¯AB(e1,e2)(x)=1IAB(e1,e2)=1IAe1(x)IBe2(x)F¯AB(e1,e2)(x)=TAB(e1,e2)=TAe1(x)TBe2(x), (32)

    and

    {T¯A¯B(e1,e2)(x)=T¯Ae1(x)T¯Be2(x)=FAe1(x)FBe2(x)I¯A¯B(e1,e2)(x)=I¯Ae1(x)I¯Be2(x)=(1IAe1(x))(1IBe2(x))F¯A¯B(e1,e2)(x)=F¯Ae1(x)F¯Be2(x)=TAe1(x)TBe2(x). (33)

    Moreover,

    (1IAe1(x))(1IBe2(x))=1IAe1(x)IBe2(x), (34)

    due to De Morgan's law of the minnorm and maxnorm. Therefore,

    (e1,e2)E×E,xX,{T¯AB(e1,e2)(x)=T¯A¯B(e1,e2)(x)I¯AB(e1,e2)(x)=I¯A¯B(e1,e2)(x)F¯AB(e1,e2)(x)=F¯A¯B(e1,e2)(x). (35)

    This section uses the operations just constructed above as the core to build the topology and related concepts on NS-sets. It is important to note that the norms used must satisfy De Morgan's law.

    Definition 14. A collection τNS(X) is NS-topology on X if it obeys the following properties:

    (a) E and XE belongs to τ,

    (b) The intersection of any finite collection of τ's elements belongs to τ,

    (c) The union of any collection of τ's elements belongs to τ.

    Then, the pair (X,τ) is a NS-topological space and each element of τ is a NS-open set.

    Example 8. Let three NS-sets K1, K2, K3 be represented in Table 12 as follows:

    Table 12.  NS-sets K1, K2, K3.
    K1 e1 e2 K2 e1 e2 K3 e1 e2
    x1 .2,.2,1 .5,.5,.7 x1 .3,.3,.9 .6,.6,.6 x1 .4,.4,.8 .7,.7,.5
    x2 .3,.3,.9 .6,.6,.6 x2 .4,.4,.8 .7,.7,.5 x2 .5,.5,.7 .8,.8,.4
    x3 .4,.4,.8 0,.9,1 x3 .5,.5,.7 .8,.8,.4 x3 .6,.6,.6 .9,.9,.3

     | Show Table
    DownLoad: CSV

    If using the minnorm xy=min{x,y}, maxnorm xy=max{x,y}, the collection τ={E,XE,K1,K2,K3} is a NS-topology.

    Theorem 7.

    (1) τ0={E,XE} is a NS-topology (anti-discrete).

    (2) τ=NS(X) is a NS-topology (discrete).

    (3) If τ1 and τ2 are two NS-topologies, τ1τ2 is a NS-topology.

    Proof. This proof focuses on the proof of Property 3 because Properties 1 and 2 are directly inferred.

    E,XEτ1;E,XEτ2E,XEτ1τ2.

    • If {Kj}n1 is a finite family of NS-sets in τ1τ2, Kiτ1 and Kiτ2 for all i. So {Kj}n1τ1 and {Kj}n1τ2. Thus {Kj}n1τ1τ2.

    • If letting {Ki|iI} be a family of NS-sets in τ1τ2, Kiτ1 and Kiτ2 for all iI. So iIKiτ1 and iIKiτ2. Therefore, iIKiτ1τ2.

    It should be noted that if τ1 and τ2 are two NS-topologies, τ1τ2 cannot be a NS-topology. Counterexamples are shown in Example 9.

    Example 9. Let three NS-sets H1, H2, H3 be represented in Table 13 as follows:

    Table 13.  NS-sets K1, K2, K3.
    H1 e1 e2 H2 e1 e2 H3 e1 e2
    x1 1,0,1 0,1,0 x1 0,1,0 1,0,1 x1 1,0,1 1,0,1
    x2 1,0,1 0,1,0 x2 0,1,0 1,0,1 x2 1,0,1 1,0,1

     | Show Table
    DownLoad: CSV

    If using the minnorm xy=xy, maxnorm xy=x+yxy and letting τ1={E,XE,H1,H2} and τ2={E,XE,H3} be two NS-topologies, the collection τ1τ2={E,XE,H1,H2,H3} is not a NS-topology due to H1H2τ1τ2, see Table 14.

    Table 14.  NS-sets H1H2.
    H1H2 e1 e2
    x1 1,0,1 1,1,0
    x2 1,0,1 1,1,0

     | Show Table
    DownLoad: CSV

    Definition 15. A NS-set ANS(X) is NS-closed set if it has the complement ¯A is a NS-open set. The symbol ¯τ is denoted as the collection of all NS-closed sets.

    Theorem 8.

    (1) E and XE belongs to ¯τ.

    (2) The union of any finite collection of ¯τ's elements belongs to ¯τ.

    (3) The intersection of any collection of ¯τ's elements belongs to ¯τ.

    Proof. These properties are directly inferred from the definitions of a NS-closed set and De Morgan's law for intersection and union.

    Definition 16. The NS-interior of a NS-set A,writtenasint(A), is the union of all NS-open subsets of A. It is considered the biggest NS-open set which is contained by A.

    Example 10. Let three NS-sets L1, L2, K be represented as follows:

    Table 15.  NS-sets L1, L2, K.
    L1 e1 e2 L2 e1 e2 K e1 e2
    x1 .7,.8,.3 .4,.5,.8 x1 .3,.2,.7 .6,.5,.2 x1 .8,.8,.3 .8,.8,.3
    x2 .5,.2,.6 .3,.4,.2 x2 .5,.8,.4 .7,.6,.8 x2 .4,.6,.5 .4,.6,.5

     | Show Table
    DownLoad: CSV

    If using the minnorm xy=max{x+y1,0}, maxnorm=min{x+y,1}, the collection τ={E,XE,L1,L2} is the NS-topology. It is easy to see that E,L1K and EL1=L1K. Therefore, int(K)=A.

    Theorem 9. A NS-set A is a NS-open set if and only if A=int(A).

    Proof. If Aτ then A is the biggest NS-open set that is contained by A. So A=int(A). Conversely, A=int(A)τ.

    Theorem 10. If A,BNS(X),

    (1) int(int(A))=int(A),

    (2) int(E)=E and int(XE)=XE,

    (3) ABint(A)int(B),

    (4) int(AB)=int(A)int(B),

    (5) int(A)int(B)int(AB).

    Proof.

    (1) Due to int(A)τ, int(int(A))=int(A).

    (2) Eτint(E)=E and XEτint(XE)=XE.

    (3) Due to AB, int(A)AB and int(B)B. Because int(B) is the biggest NS-open set contained in B, int(A)int(B).

    (4) Since int(A)τ and int(B)τ, then int(A)int(B)τ. It is known that int(A)A and int(B)B, so int(A)int(B)AB. Moreover, int(AB) is the biggest NS-open set contained in AB. Therefore, int(A)int(B)int(AB).

    (5) Since int(AB)AB, so int(AB)A and int(AB)B. Therefore, int(AB)int(A) and int(AB)int(B) or int(AB)int(A)int(B).

    Moreover,

    {int(A)int(B)int(A)Aint(A)int(B)int(B)Bint(A)int(B)AB

    and int(AB) is the biggest NS-open set contained in AB, so

    int(A)int(B)int(AB).

    Thus, int(AB)=int(A)int(B).

    Definition 17. The NS-closure of a NS-set A,writtenascl(A), is the intersection of all NS-closed supersets of A. The cl(A) is the smallest NS-closed set which contains A.

    Example 11. For the NS-topology τ given in Example 10, let NS-set H be represented in Table 16 as follows:

    Table 16.  NS-sets H.
    H e1 e2 ¯L1 e1 e2 e1 e2
    x1 .2,.2,.8 .2,.2,.8 x1 .3,.2,.7 .8,.5,.4 .7,.8,.3 .2,.5,.6
    x2 .3,.4,.8 .3,.4,.8 x2 .6,.8,.5 .2,.6,.3 .4,.2,.5 .8,.4,.7

     | Show Table
    DownLoad: CSV

    It is easy to see that ¯E=XE, ¯XE=E. So E, XE, ¯L1, ¯L2 are all NS-closed sets. Since HXE, cl(H)=L2.

    Theorem 11. A NS-set A is a NS-closed set if and only if A=cl(A).

    Proof. Let A be a NS-closed set. Because AA and cl(A) is the smallest NS-closed set that contains A, cl(A)A. Therefore, A=cl(A). Conversely, if A=cl(A) then A is a NS-closed set.

    Theorem 12. If A,BNS(X),

    (1) cl(cl(A))=cl(A),

    (2) cl(E)=E and cl(XE)=XE,

    (3) ABcl(A)cl(B),

    (4) cl(AB)cl(A)cl(B),

    (5) cl(AB)=cl(A)cl(B).

    Proof.

    (1) Directly inferring from Theorem 9.

    (2) Directly inferring from Definition 14 and Theorem 9.

    (3) Since ABcl(B) and cl(A) is the smallest NS-closed set containing A, cl(A)cl(B).

    (4) Since ABAcl(A) and ABBcl(B), ABcl(A)cl(B). Therefore, cl(AB)cl(A)cl(B).

    (5) It is easy to see that AABcl(AB), BABcl(AB), cl(A) is the smallest NS-closed set that contains A, and cl(B) is the smallest NS-closed set that containing B. So cl(A)cl(AB) and cl(B)cl(AB). Therefore, cl(A)cl(B)cl(AB).

    Moreover, since Acl(A) and Bcl(B), ABcl(A)cl(B). Therefore, cl(AB)cl(A)cl(B).

    Thus, cl(AB)=cl(A)cl(B).

    Theorem 13. If A,BNS(X),

    (1) ¯int(A)=cl(¯A),

    (2) ¯cl(A)=int(¯A).

    Proof.

    (1) Because

    int(A)=iI{Hiτ:HiA},
    ¯int(A)=¯iI{Hiτ:HiA}=iI{¯Hi¯τ:¯Hi¯A}=cl(¯A). (36)

    (2) Because

    cl(A)=iI{Hi¯τ:HiA},
    ¯cl(A)=¯[iI{Hi¯τ:HiA}]=iI{¯Hiτ:¯Hi¯A}=int(¯A). (37)

    Definition 18. The NS-boundary of a NS-set A,writtenasA, is the intersection of the NS-closure of A and the NS-closure of ¯A.

    Example 12. For the NS-topology τ given in Example 10 and the NS-set H given in Example 11, the complement of H is represented in Table 17 as follows:

    Table 17.  NS-sets ¯H.
    ¯H e1 e2
    x1 .8,.8,.2 .8,.8,.2
    x2 .8,.6,.3 .8,.6,.3

     | Show Table
    DownLoad: CSV

    It is easy to see that cl(H)=XE and cl(¯H)=XE. So H=XEXE=XE.

    Theorem 14. If ANS(X),

    (1) A=cl(A)¯int(A),

    (2) int(A)A=E,

    (3) A=E if and only if A is a NS-open and NS-closed set.

    Proof.

    (1) A=cl(A)cl(¯A) = cl(A)¯int(A) due to Theorem 13.

    (2) It is easy to see that

    int(A)(A)=int(A)cl(A)cl(¯A)=int(A)cl(A)¯int(A)
    =int(A)¯int(A)cl(A)=E

    due to Theorem 13.

    (3) Since

    (A)=cl(A)cl(¯A)=cl(A)¯int(A)=E,

    cl(A)int(A)E. So Acl(A)int(A)A. Therefore, A=cl(A)=int(A) or A is a NS-open and NS-closed set.

    Conversely, if A is a NS-open and NS-closed set, A=int(A) and A=cl(A). Therefore,

    (A)=cl(A)cl(¯A)=cl(A)¯int(A)=cl(A)¯cl(A)=E.

    Definition 19.

    a. The NS-open set M is regular if M=int(cl(M)).

    b. The NS-closed set M is regular if M=cl(int(M)).

    Theorem 15. If M,NNS(X),

    (1) If M is a NS-closed set, int(M) is a regular NS-open set.

    (2) If M is a NS-open set, cl(M) is a regular NS-closed set.

    (3) If M and N are two regular NS-open sets, MNcl(M)cl(N).

    (4) If M and N are two regular NS-closed sets, MNint(M)int(N).

    (5) If M is a regular NS-closed set, ¯M is a regular NS-open set.

    (6) If M is a regular NS-open set, ¯M is a regular NS-closed set.

    Proof.

    (1) If M is a NS-closed set,

    int(M)Mcl[int(M)]cl(M)=Mint[cl(int(M))]int(M). (38)
    int(M)cl(int(M))int(int(M))=int(M)int[cl(int(M))]. (39)

    Therefore, int(M)=int[cl(int(M))] or int(M) is regular.

    (2) If M is a NS-open set,

    int(cl(M))cl(M)cl(int(cl(M)))cl(cl(M))=cl(M). (40)

    Because int(M)=M,

    Mcl(M)int(M)=Mint(cl(M))cl(M)cl(int(cl(M))). (41)

    Therefore, cl(M)=cl(int(cl(M))) or cl(M) is regular.

    (3) Clearly, MNcl(M)cl(N) and int(cl(M))=M, int(cl(N))=N due to M,N are regular. Conversely,

    cl(M)cl(N)int(cl(M))=Mint(cl(N))=NMN.

    (4) Clearly, MNint(M)int(N) and M=cl(int(M)), N=cl(int(N)) due to M,N are regular. Conversely,

    int(M)int(N)cl(int(M))=Mcl(int(N))=NMN.

    (5) If M is a regular NS-open set, M=int(cl(M)). So

    cl(int(¯M))=cl(¯cl(M))=¯int(cl(M))=¯M.

    Therefore, ¯M is a regular NS-closed set.

    (6) Similarly, if M is a regular NS-closed set, int(cl(¯M))=¯M. So ¯M is a regular NS-open set.

    Theorem 16. Let τ={Ki:iI} be NS-topology on X where

    Ki={(e,xTKie(x),IKie(x),FKie(x)):eE,xX}. (42)

    Three collections

    T=(Ti)iI={(e,x,TKie(x)):eE,xX}, (43)
    I=(Ii)iI={(e,x,IKie(x)):eE,xX}, (44)
    F=(Fi)iI={(e,x,1FKie(x)):eE,xX}, (45)

    are the fuzzy soft topologies on X.

    Proof.

    Eτ˜T,˜I,˜F.

    XEτ˜XT;˜XI;˜XF.

    • Let (Ti)iI be a family of fuzzy soft sets in T, (Ii)iI be a family of fuzzy soft sets in I, and (Fi)iI be a family of fuzzy soft sets in F. They make a family of NS-sets {Ki:iI} where

    Ki={(e,xTKie(x),IKie(x),FKie(x)):eE,xX}τ. (46)

    So iIKiτ or

    iIKi={(e,xiI{TKie(x)},iI{IKie(x)},iI{FKie(x)}):eE,xX}τ. (47)

    Therefore,

    {[iI{TKie(x)}:xX]eE}=˜iI{TKie(X):eE}T, (48)
    {[iI{TKie(x)}:xX]eE}=˜iI{IKie(X):eE}I, (49)
    {[iI{FKie(x)}:xX]eE}C
    ={[1iI{FKie(x)}:xX]eE}
    ={[iI{(1FKi(e)(a))}:aX]eE}
    =˜iI{(IKi(e)(X))CeE}F. (50)

    • Let {TjT}n1, {IjI}n1,{FjF}n1 be finite families of fuzzy soft sets on X and satisfy

    Kj={(e,xTKje(x),IKje(x),FKje(x)):eE,xX}τ. (51)

    So, we have n1Kjτ, i.e.,

    n1Kj={(e,x{TKie(x)}n1,{IKie(x)}n1,{FKie(x)}n1):eE,xX}. (52)

    Hence,

    [{{TKie(x)}n1:xX}eE]=˜n1{[TKie(X)]eE}T, (53)
    [{{IKie(x)}n1:xX}eE]=˜n1{[IKie(X)]eE}I, (54)
    {[{FKie(x)}n1:xX]eE}C
    ={[1{FKie(x)}n1:xX]eE}
    ={[{1FKie(x)}n1:xX]eE}
    =˜n1{[IKie(X)]eE}F. (55)

    In the general case, the opposite of Theorem 16 is not true. This is demonstrated through a counterexample, as shown in Example 13.

    Example 13. Let four NS-sets H1, H2, and H3be represented in Table 18 as follows:

    Table 18.  NS-sets H1, H2, and H3.
    e1 e2
    H1 x1 .25,.25,.75 .25,.25,.75
    x2 13,13,13 13,13,13
    H2 x1 .5,.75,.5 .5,.75,.5
    x2 13,23,23 13,23,23
    H3 x1 .75,.5,.25 .75,.5,.25
    x2 23,13,13 23,13,13
    H1H2 x1 .75,1,.25 .75,1,.25
    x2 23,1,0 23,1,0

     | Show Table
    DownLoad: CSV

    If using the minnorm xy=max{x+y1,0}, maxnorm=min{x+y,1}, three collections defined in Table 19 as follows are the fuzzy soft topologies on X.

    Table 19.  NS-sets T, I, and F.
    e1 e2
    T ˜ 0,0 0,0
    ˜X 1,1 1,1
    T1 .25,13 .25,13
    T2 .5,13 .5,13
    T3 .75,23 .75,23
    I ˜ 0,0 0,0
    ˜X 1,1 1,1
    I1 .25,13 .25,13
    I2 .75,23 .75,23
    I3 .5,13 .5,13
    F ˜ 0,0 0,0
    ˜X 1,1 1,1
    F1 .25,13 .25,13
    F2 .5,13 .5,13
    F3 .75,23 .75,23

     | Show Table
    DownLoad: CSV

    The T,I,F are fuzzy soft topologies on X, but τ={E,XE,H1,H2,H3} is not a NS- topology on X because H1H2τ.

    Theorem 17. Let three collections

    T=(Ti)iI={(e,x,TKie(x)):eE,xX}, (56)
    I=(Ii)iI={(e,x,IKie(x)):eE,xX}, (57)
    F=(Fi)iI={(e,x,1FKie(x)):eE,xX}, (58)

    be the fuzzy soft topologies on X. Let τ={Ki:iI} where

    Ki={(e,xTKie(x),IKie(x),FKie(x)):eE,xX}. (59)

    If for all l,m,n, we have

    TlTm=Tn{IlIm=InFlFm=Fn, (60)
    TlTm=Tn{IlIm=InFlFm=Fn. (61)

    Then τ is the NS-topology on X.

    Proof.

    • Obviously, E,XEτ.

    • Let {Ki:iI}τ be a family of NS-sets on X. We have {Ti},{Ii},{Fi} are families of fuzzy soft sets on X. So,

    n0I,Tn0=iITiT,In0=iIIiI,Fn0=iIFiF. (62)

    Thus, iIKi=Tn0τ.

    • Let {Kjτ}n1 be a finite family of NS-sets on X. We have {Tj}n1, {Ij}n1,{Fj}n1as finite families of fuzzy soft sets on X. So,

    m0I,Tm0=n1TjT,Im0=n1IjI,Fm0=n1FjF, (63)

    Thus, n1Kjτ.

    Theorem 18. Let τ={Ki:iI} be the NS-topology on X where

    Ki={(e,xTKie(x),IKie(x),FKie(x)):eE,xX}. (64)

    For each eE, three collections

    Te=(Tei)iI={x,TKie(x):xX}, (65)
    Ie=(Iei)iI={x,IKie(x):xX}, (66)
    Fe=(Fei)iI={x,1FKie(x),xX}, (67)

    are the fuzzy topologies on X.

    Proof. It can be implied from Theorem 17.

    In the general case, the opposite of Theorem 18 is not true. This is demonstrated through the counterexample shown in Example 14.

    Example 14. We return to Example 12 with the same hypothesis. Then,

    Te1={(0,0),(1,1),(.25,13),(.5,13),(.75,23)}, (68)
    Ie1={(0,0),(1,1),(.25,13),(.75,23),(.5,13)}, (69)
    Fe1={(0,0),(1,1),(.25,13),(.5,13),(.75,23)}, (70)

    are fuzzy topologies on X. Similarly,

    Te2={(0,0),(1,1),(.25,13),(.5,13),(.75,23)}, (71)
    Ie2={(0,0),(1,1),(.25,13),(.5,13),(.75,23)}, (72)
    Fe2={(0,0),(1,1),(.25,13),(.5,13),(.75,23)}, (73)

    are also fuzzy topologies, but τ={E,XE,H1,H2,H3} is not a NS-topology on X because K1K2τ.

    In this paper, two novel norms are proposed to serve as the core for determining operations on NS-sets. These operations are used to construct the topology and related concepts such as open set, closed set, interior, closure, and regularity. Another highlight of this work is demonstrating the relationship between the topologies on NS-sets and fuzzy soft sets. The topology on NS-sets can parameterize the topologies on fuzzy soft sets, but the reverse is not guaranteed. This work's advantage is the structure's logic is presented with well-defined concepts and convincingly proven theorems.

    Determining these concepts in a novel way enables a variety of methods for studying NS-sets, and offers a unique opportunity for future research and development in this field. Such research could focus on separation axioms, continuity, compactness, and paracompactness on NS-sets. Moreover, the relationship between topology on hybrid structure, NS-sets, and component structures, neutrosophic sets and soft sets, is also of research interest. In addition, applications of neutrosophic soft topological spaces can be investigated to handle decision-making problems.

    Furthermore, we are also turning our interests to building topology on a new type of set, neutrosophic fuzzy sets. We believe these results will be helpful for future studies on neutrosophic fuzzy topology to develop a general framework for practical applications. These issues present opportunities but also challenges for researchers interested in the field of fuzzy theory.

    The authors declare that they have no competing interests in this paper.



    [1] O. Berman, Z. Drezner, A new formulation for the conditional p-median and p-venter problem, Oper. Res. Lett., 36 (2008), 481-483. doi: 10.1016/j.orl.2008.02.001
    [2] G. Buttazzo, E. Oudet, E. Stepanov, Optimal transportation problems with free dirichlet regions, In: G. dal Maso, F. Tomarelli, Editors, Variational Methods for Discontinuous Structures, Birkhäuser Verlag, Basel, 51 (2002), 41-65.
    [3] G. Buttazzo, A. Pratelli, S. Solimini, et al. Optimal Urban Networks via Mass Transportation, Springer, Berlin, 2009.
    [4] Z. Drezner, The p-centre problem - heuristic and optimal algorithms, J. Oper. Res. Soc., 35 (1984), 741-748.
    [5] A. Heppes, H. Melissen, Covering a rectangle with equal circles, Period. Math. Hung., 34 (1997), 65-81. doi: 10.1023/A:1004224507766
    [6] M. E. Johnson, L. M. Moore, D. Ylvisaker, Minimax and maximin distance designs, J. Stat. Plan. Infer., 26 (1990), 131-148. doi: 10.1016/0378-3758(90)90122-B
    [7] W. C. Ke, B. H. Liu, M. J. Tasi, Efficient algorithm for constructing minimum size wireless sensor networks to fully cover critical square grids, IEEE T. Wirel. Commun., 10 (2011), 1154-1164. doi: 10.1109/TWC.2011.021611.100123
    [8] J. B. M. Melissen, P. C. Schuur, Improved coverings of a square with six and eight equal circles, Electron. J. Comb., 3 (1996), 1-10.
    [9] F. Morgan, R. Bolton, Hexagonal economic regions solve the location problem, Am. Math. Mon., 109 (2002), 165-172. doi: 10.1080/00029890.2002.11919849
    [10] K. J. Nurmela, P. R. J. Östergård, Packing up to 50 equal circles in a square, Discrete Comput. Geom., 18 (1997), 111-120. doi: 10.1007/PL00009306
    [11] K. J. Nurmela, P. R. J. Östergård, Covering a square with up to 30 circles, Research report, Helsinki University of Technology, Laboratory for Theoretical Computer Science, 2000.
    [12] J. Schaer, The densest packing of 9 circles in a square, Can. Math. Bull., 8 (1965), 273-277. doi: 10.4153/CMB-1965-018-9
    [13] D. Spernjaka, A. K. Prasada, S. G. Advani, Experimental investigation of liquid water formation and transport in a transparent single-serpentine PEM fuel cell, J. Power Sources, 170 (2007), 334-344. doi: 10.1016/j.jpowsour.2007.04.020
    [14] Specht, The best known packings of equal circles in a square, Available from: http://hydra.nat.uni-magdeburg.de/packing/csq/csq.html, accessed November 1, 2012.
    [15] T. Tarnai, Z. Gaspar, Covering a square by equal circles, Elem. Math., 50 (1995), 167-170.
    [16] Y. G. Stoyan, V. M. Patsuk, Covering a compact polygonal set by identical circles, Comput. Optim. Appl., 46 (2010), 75-92. doi: 10.1007/s10589-008-9191-8
    [17] A. E. Xavier, A. A. F. de Oliviera, Optimal covering of plane domains by circles via hyperbolic smoothing, J. Global Optim., 31 (2005), 493-504. doi: 10.1007/s10898-004-0737-8
  • This article has been cited by:

    1. Adem Yolcu, Aysun Benek, Taha Yasin Öztürk, A new approach to neutrosophic soft rough sets, 2023, 0219-1377, 10.1007/s10115-022-01824-z
    2. Quang-Thinh Bui, My-Phuong Ngo, Vaclav Snasel, Witold Pedrycz, Bay Vo, Information measures based on similarity under neutrosophic fuzzy environment and multi-criteria decision problems, 2023, 122, 09521976, 106026, 10.1016/j.engappai.2023.106026
    3. Bing Chen, Xiao Long Xin, Xiao Fei Yang, Distance functions and filter topological residuated lattices1, 2024, 10641246, 1, 10.3233/JIFS-238147
    4. Nechervan B. Ibrahim, Alias B. Khalaf, On maximal block topological space, 2023, 45, 10641246, 8541, 10.3233/JIFS-223749
    5. Aysun Benek, Taha Yasin Öztürk, Adem Yolcu, 2024, chapter 4, 9798369320853, 67, 10.4018/979-8-3693-2085-3.ch004
  • Reader Comments
  • © 2019 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(5592) PDF downloads(341) Cited by(1)

Figures and Tables

Figures(3)

Other Articles By Authors

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog