Show Summary Details
More options …

# Open Mathematics

### formerly Central European Journal of Mathematics

Editor-in-Chief: Gianazza, Ugo / Vespri, Vincenzo

1 Issue per year

IMPACT FACTOR 2017: 0.831
5-year IMPACT FACTOR: 0.836

CiteScore 2017: 0.68

SCImago Journal Rank (SJR) 2017: 0.450
Source Normalized Impact per Paper (SNIP) 2017: 0.829

Mathematical Citation Quotient (MCQ) 2017: 0.32

ICV 2017: 161.82

Open Access
Online
ISSN
2391-5455
See all formats and pricing
More options …
Volume 15, Issue 1

# Dynamics for a discrete competition and cooperation model of two enterprises with multiple delays and feedback controls

Lin Lu
• Corresponding author
• Research Center of Modern Enterprise Management of Guilin University of Technology, Guilin University of Technology, Guilin 541004, China
• Guilin University of Aerospace Technology, Guilin 541004, China
• Other articles by this author:
/ Yi Lian
• Research Center of Modern Enterprise Management of Guilin University of Technology, Guilin University of Technology, Guilin 541004, China
• Other articles by this author:
/ Chaoling Li
• Research Center of Modern Enterprise Management of Guilin University of Technology, Guilin University of Technology, Guilin 541004, China
• Guilin University of Aerospace Technology, Guilin 541004, China
• Email
• Other articles by this author:
Published Online: 2017-03-13 | DOI: https://doi.org/10.1515/math-2017-0023

## Abstract

This paper is concerned with a competition and cooperation model of two enterprises with multiple delays and feedback controls. With the aid of the difference inequality theory, we have obtained some sufficient conditions which guarantee the permanence of the model. Under a suitable condition, we prove that the system has global stable periodic solution. The paper ends with brief conclusions.

MSC 2010: 34K20; 34C25; 92D25

## 1 Introduction

It is known that the coexistence of species has become one of interesting subjects in mathematical ecology. In the past few decades, permanence dynamics of species have received great attention and have been investigated in a number of notable work. For example, Wang and Huang [1] analyzed permanence of a predator-prey model with harvesting predator. Mukherjee [2] addressed the permanence and global attractivity for facultative mutualism predator-prey model, Zhao and Jiang [3] considered the permanence and extinction for Lotka-Volterra model, Teng et al. [4] established the permanence criteria for a delayed discrete species systems, Liu et al. [5] studied the permanence and periodic solutions for reaction-diffusion food-chain system with impulsive effect. For more detailed research about this topic, one can see [623]. In real life, the co-existence and stability of enterprise clusters has become one of the most prevalent phenomena in our society. Thus it is important for us to study the permanence and global attractivity of enterprise clusters. However, there are few papers that consider this topic. We think that this study on the dynamics of enterprise clusters has wide application in economic performance and so on.

In 2006, Tian and Nie [24] investigated the following competition and cooperation model of two enterprises $dx1(t)dt=r1(t)x1(t)1−x1(t)K−α(x2(t)−c2)2K,dx2(t)dt=r2(t)x2(t)1−x2(t)K−β(x1(t)−c1)2K,$(1)

where x1(t), x2(t) represent the output of enterprises A and B, r1, r2 are the intrinsic growth rate, K denotes the carrying capacity of mark under nature unlimited conditions, α, β are the competitive parameters of two enterprises, c1, c2 are the initial production of two enterprises. Letting ${a}_{1}=\frac{{r}_{1}}{K},\phantom{\rule{thinmathspace}{0ex}}{a}_{2}=\frac{{r}_{2}}{K},\phantom{\rule{thinmathspace}{0ex}}{b}_{1}=\frac{{r}_{1}\alpha }{K},\phantom{\rule{thinmathspace}{0ex}}{b}_{2}=\frac{{r}_{2}\beta }{K},$ then system (1) becomes $dx1(t)dt=x1(t)[r1−a1x1(t)−b1(x2(t)−c2)2],dx2(t)dt=x2(t)[r2−a2x2(t)+b2(x1(t)−c1)2].$(2)

Considering the effect of time delay, Liao et al. [25] modified system (2) as follows: $dx1(t)dt=x1(t)[r1−a1x1(t−τ1)−b1(x2(t−τ2)−c2)2],dx2(t)dt=x2(t)[r2−a2x2(t−τ1)+b2(x1(t−τ2)−c1)2],$(3) where τ1 is nonnegative constant which stands for the gestation periodic of production for two enterprises, τ2 in the first equation of the system (3) stands for the block delay of enterprise B to A, and τ2 in the second equation of the system (3) stands for the promoting delay of enterprise A to B. By regarding the two delays τ1 and τ2 as bifurcation parameters, Liao et al. [25] discussed the effect of different delays on the dynamical behavior of system (3). If τ1 = τ2 = τ, then system (3) becomes $dx1(t)dt=x1(t)[r1−a1x1(t−τ)−b1(x2(t−τ)−c2)2],dx2(t)dt=x2(t)[r2−a2x2(t−τ)+b2(x1(t−τ)−c1)2],$(4) By choosing the time delay τ as bifurcation parameter, Liao et al. [26] focused on the stability and Hopf bifurcation properties of system (4).

Li and Zhang [27] focused on (2) with nonconstant coefficients, which takes the following from: $dx1(t)dt=x1(t)[r1(t)−a1(t)x1(t)−b1(t)(x2(t)−c2(t))2],dx2(t)dt=x2(t)[r2(t)−a2(t)x1(t)+b2(t)(x1(t)−c1(t))2].$(5)

Using the continuation theorem of coincidence degree theory and differential inequality theory, Xu [28] established some sufficient criteria to guarantee the existence of periodic solutions of (5).

Considering that the change of environment, the output of enterprises A and B usually change rapidly, Xu and Shao [29] considered the existence and global attractivity of periodic solution for the following enterprise clusters model with impulse and varying coefficients $dx1(t)dt=x1(t)[r1(t)−a1(t)x1(t)−b1(t)(x2(t)−c2(t))2],t≠tk,dx2(t)dt=x2(t)[r2(t)−a2(t)x1(t)+b2(t)(x1(t)−c1(t))2],t≠tk,Δi(tk)=xi(tk+)−xi(tk−)=−γik(tk),i=1,2,k=1,2,⋯,q,$(6) where ${\mathrm{\Delta }}_{i}\left({t}_{k}\right)={x}_{i}\left({t}_{k}^{+}\right)-{x}_{i}\left({t}_{k}^{-}\right)$ are the impulses at moments tk and t1 < t2 < ⋯ is a strictly increasing sequence such that limk→+∞ tk = +∞ and q is a positive integer. Applying coincidence degree theory, Xu and Shao [29] obtained a set of sufficient criteria to ensure the existence of at least a positive periodic solution, and by constructing a Lyapunov functional, they established a sufficient condition to guarantee the uniqueness and global attractivity of the positive periodic solution for (6).

A lot of researchers [3041] think that discrete systems are far better in depicting the dynamical behavior than continuous ones. In addition, discrete systems also play an important role in computer simulations for continuous systems. Considering the unpredictable forces, we know that coefficients of competition and cooperation systems of two enterprises change with time. Inspired by the discussion above, we modify (5) as follows $x1(n+1)=x1(n)exp⁡{r1(n)−a1(n)x1(n−τ1(n))−b1(n)(x2(n−τ2(n))−c2(n))2−β1(n)U1(n)},x2(n+1)=x2(n)exp⁡{r2(n)−a2(n)x2(n−τ1(n))+b2(n)(x1(n−τ2(n))−c1(n))2−β2(n)u2(n)},Δu1(n)=−γ1(n)u1(n)+η1(n)x1(n),Δu2(n)=−γ2(n)u2(n)+η2(n)x2(n),$(7) where x1(n) and x2(n) denote the output of enterprises A and B at the generation, respectively, and ui(n)(i = 1, 2) is the control variable. r1(n), r2(n), a1(n), a2(n), b1(n), b2(n), c1(n), c2(n), τ1(n) and τ2(n) are bounded nonnegative sequences. To the authors' knowledge, it is the first time one deals with system (7) with feedback control. For more related work, one can see [4244].

We assume that $(H1)0

Here, for any bounded sequence {f(n)}, fu = supnN{f(n)} and fl = infnN{f(n)}.

Let τ = supnZ{τi}(n)}, $\stackrel{~}{\tau }$ = infnZ{τi(n)}, i = 1, 2. The initial conditions of (7) are $xi(θ)=φi(θ)≥0,θ∈N[−τ,0]={−τ,−τ+1,⋯,0},φi(0)>0.$(8)

It is not difficult to see that solutions of (7) and (8) are well defined for all n ≥ 0 and satisfy $xi(n)>0,forn∈Z,i=1,2.$

The remainder of the article is organized as follows: in Section 2, some definitions and lemmas are presented and the permanence of (7) is considered. In Section 3, the existence and stability of a unique globally attractive positive periodic solution of the model are investigated. In Section 4, two examples are given to illustrate correctness of our obtained analytical results in Section 2 and Section 3. Brief conclusions are drawn in Section 5.

## 2 Permanence

In this section, we list several definitions and lemmas.

#### Definition 2.1

System (7) is permanent if there are positive constants M and m such that for each positive solution (x1(n), x2(n), u1(n), u2(n)) of system (7) satisfies $m≤limn→+∞infxi(n)≤limn→+∞supxi(n)≤M(i=1,2),m≤limn→+∞infui(n)≤limn→+∞supui(n)≤M(i=1,2).$

Let us consider the following model: $N(n+1)=N(n)exp⁡(a(n)−b(n)N(n)),$(9) where {a(n)} and {b(n)} are strictly positive sequences of real numbers defined for nN = {0, 1, 2, } and 0 < alau, 0 < blbu. Similarly to the proofs of Propositions 1 and 3 in [41], we can obtain the following Lemma 2.2.

#### Lemma 2.2

Any solution of system (9) with initial condition N(0) > 0 satisfies $m≤limn→+∞infN(n)≤limn→+∞supN(n)≤M,$ where $M=1blexp⁡(au−1),m=albuexp⁡(al−buM).$

Let consider the following equation $y(n+1)=Ay(n)+B,n=1,2,⋯,$(10) where A and B are positive constants. Following Theorem 6.2 of Wang and Wang [45, page 125], we have the following Lemma 2.3.

#### Lemma 2.3

([45]). If|A| < 1, then for any initial value y(0), there exists a unique solution y(n) of (10) which can be expressed as follows: $y(n)=An(y(0)−y∗)+y∗,$ where ${y}^{\ast }=\frac{B}{1-A}.$ Thus, for any solution {y(n)} of system (10), limn→+∞y(n) = y*.

#### Lemma 2.4

([45]). Let $n\in {N}_{{n}_{0}}^{+}=\left\{{n}_{0},{n}_{0}+1,\cdots ,{n}_{0}+l,\cdots \right\},r\ge 0.$ For any fixed n, g(n, r) is a nondecreasing function with respect to r, and for nn0, then $y(n+1)≤g(n,y(n)),u(n+1)≥g(n,u(n)).$

If y(n0) ≤ u(n0), then y(n) ≤ u(n) for all nn0.

#### Proposition 2.5

Let ε > 0 be any constant. If(H1) holds, then $limn→+∞supxi(n)≤Mi,limn→+∞supui(n)≤Ui,i=1,2,$ where $M1=1r1lexp⁡{r1u(τ+1)−1},Ui=ηiuMiγil(i=1,2),M2=1a2lexp⁡{(r2u+b2u(M1+ϵ+c1u)2)(τ+1)−1}.$

#### Proof

Let (x1(n), x2(n), u1(n), u2(n)) be any positive solution of system (7) with the initial condition (x1(0), x2(0), u1(0), u2(0)) . It follows from the first equation of system (7) that $x1(n+1)≤x1(n)exp⁡{r1(n)}.$(11)

Let x1(n) = exp{y1(n)}, then (11) is equivalent to $y1(n+1)−y1(n)≤r1(n).$(12)

Summing both sides of (12) from nτ1(n) to n−1, we have $∑j=n−τ1(n)n−1(y1(j+1)−y1(j))≤∑j=n−τ1(n)n−1r1(j)≤r1uτ,$ which leads to $y1(n−τ1(n))≥y1(n)−r1uτ.$(13)

Then $x1(n−τ1(n))≥x1(n)exp⁡{−r1uτ}.$(14)

Substituting (14) into the first equation of system (7) gives $x1(n+1)≤x1(n)exp⁡{r1(n)−r1(n)exp⁡{−r1uτ}x1(n)}.$(15)

It follows from (15) and Lemma 2.2 that $limn→+∞supx1(n)≤1r1lexp⁡{r1u(τ+1)−1}:=M1.$(16)

For any positive constant ε > 0, it follows (16) that there exists a N1 > 0 such that for all n > N1 + τ $x1(n)≤M1+ε.$(17)

For any positive constant ε > 0 and for all n > N1 + τ, by the second equation of system (7), we get $x2(n+1)≤x2(n)exp⁡{r2(n)+b2(n)(M1+ε+c1(n))2}.$(18)

Let x2(n) = exp{y2(n)}, then (18) is equivalent to $y2(n+1)−y2(n)≤r2(n)+b2(n)(M1+ε+c1(n))2.$(19)

Summing both sides of (19) from nτ1(n) to n − 1, we have $∑j=n−τ1(n)n−1(y2(j+1)−y2(j))≤∑j=n−τ1(n)n−1[r2(n)+b2(n)(M1+ε+c1(n))2]≤[r2u+b2u(M1+ε+c1u)2]τ,$ which leads to $y2(n−τ1(n))≥y2(n)−[r2u+b2u(M1+ε+c1u)2]τ.$(20)

Then $x2(n−τ1(n))≥x2(n)exp⁡{−[r2u+b2u(M1+ε+c1u)2]τ}.$(21)

By substituting (21) into the second equation of system (7) we have $x2(n+1)≤x2(n)exp⁡{r2(n)+b2(n)(M1+ε+c1(n))2−a2(n)exp⁡{−[r2u+b2u(M1+ε+c1u)2]τ}x2(n)}.$(22)

It follows from (22) and Lemma 2.2 that $limn→+∞supx2(n)≤1a2lexp⁡{(r2u+b2u(M1+ε+c1u)2)(τ+1)−1}:=M2.$(23)

For any positive constant ε > 0, it follows (23) that there exists a N2 > N1 + τ such that for all n > N2 + τ $x2(n)≤M2+ε.$(24)

In view of the third and fourth equations of the system (7), we can obtain $Δu1(n)≤−γ1(n)u1(n)+η1(n)(M1+ε),$(25) $Δu2(n)≤−γ2(n)u2(n)+η2(n)(M2+ε).$(26)

Then $u1(n+1)≤(1−γ1l)u1(n)+η1u(M1+ε),$(27) $u2(n+1)≤(1−γ2l)u2(n)+η2u(M2+ε).$(28)

Applying Lemmas 2.2 and 2.4, we immediately get $limn→+∞supu1(n)≤η1u(M1+ε)γ1l,$(29) $limn→+∞supu2(n)≤η2u(M2+ε)γ2l.$(30)

By setting ε → 0, we have $limn→+∞supu1(n)≤η1uM1γ1l:=U1,$(31) $limn→+∞supu2(n)≤η2uM2γ2l:=U2.$(32)

This completes the proof of Proposition 2.5.

#### Theorem 2.6

Let M1, M2, U1 and U2 be defined by (16), (23), (31) and (32), respectively. Assume that (H1) and (H2) $\begin{array}{}{r}_{1}^{l}>{b}_{1}^{u}\left({M}_{2}+{c}_{2}^{u}{\right)}^{2}+{\beta }_{1}{U}_{1},{r}_{2}^{l}>{\beta }_{2}^{u}{U}_{2}\end{array}$ hold, then system (7) is permanent.

#### Proof

By applying Proposition 2.5, we easily see that to end the proof of Theorem 2.6, it is enough to show that under the conditions of Theorem 2.6, $limn→+∞infx1(n)≥m1,limn→+∞infx2(n)≥m2,limn→+∞infu1(n)≥v1,limn→+∞infu2(n)≥v2.$

In view of Proposition 2.5, for all ε > 0, there exists a N3 > 0, N3N, for all n > N3, $x1(n)≤M1+ε,x2(n)≤M2+ε,u1(n)≤U1+ε,u2(n)≤U2+ε.$(33)

It follows from the first equation of system (7) and (33) that $x1(n+1)≥x1(n)exp⁡{r1l−a1u(M1+ε)−b1u(M2+ε+c2u)2−β1u(U1+ε)},$ for all n > N3 + τ.

Let x1(n) = exp{y1(n)}, then (34) is equivalent to $y1(n+1)−y1(n)≥r1l−a1u(M1+ε)−b1u(M2+ε+c2u)2−β1u(U1+ε).$(34)

Summing both sides of (34) from nτ1(n) to n − 1 leads to $∑j=n−τ1(n)n−1(y1(j+1)−y1(j))≥∑j=n−τ1(n)n−1[r1l−a1u(M1+ε)−b1u(M2+ε+c2u)2−β1u(U1+ε)]≥[r1l−a1u(M1+ε)−b1u(M2+ε+c2u)2−β1u(U1+ε)]τ~.$

Then $y1(n−τ1(n))≤y1(n)−[r1l−a1u(M1+ε)−b1u(M2+ε+c2u)2−β1u(U1+ε)]τ~.$

Thus $x1(n−τ1(n))≤x1(n)exp⁡{−[r1l−a1u(M1+ε)−b1u(M2+ε+c2u)2−β1u(U1+ε)]τ~}.$(35)

Substituting (33) and (35) into the first equation of (7), we have $x1(n+1)≥x1(n)exp⁡{r1l−b1u(M2+ε+c2u)2−β1u(U1+ε)−a1ux1(n)×exp⁡{−[r1l−a1u(M1+ε)−b1u(M2+ε+c2u)2−β1u(U1+ε)]τ~}}$(36) for all n > N3 + τ.

By applying Lemmas 2.2 and 2.4, we immediately obtain $limn→+∞infx1(n)≥m1ε,$(37) where $m1ε=r1l−b1u(M2+ε+c2u)2−β1u(U1+ε)a1uexp⁡{−[r1u−a1u(M1+ε)−b1u(M2+ε+c2u)2−β1u(U1+ε)]τ~}×exp⁡{r1l−b1u(M2+ε+c2u)2−β1u(U1+ε)−a1uexp⁡{−[r1u−a1u(M1+ε)−b1u(M2+ε+c2u)2−β1u(U1+ε)]τ~}M1}.$

Setting ε → 0 in (37), then $limn→+∞infx1(n)≥m1,$(38) where $m1=r1l−b1u(M2+c2u)2−β1uU1a1uexp⁡{−[r1u−a1uM1−b1u(M2+c2u)2−β1uU1]τ~}×exp⁡{r1l−b1u(M2+c2u)2−β1uU1−a1uexp⁡{−[r1u−a1uM1−b1u(M2+c2u)2−β1uU1]τ~}M1}.$

By the second equation of system (7) and (33), we can obtain $x2(n+1)≥x2(n)exp⁡{r2(n)−a2(n)x2(n−τ1(n))−β2(n)u2(n)}≥x2(n)exp⁡{r2l−a2u(M2+ε)−β2u(U2+ε)}$(39) for all n > N3 + τ.

Let x2(n) = exp{y2(n)}, then (39) is equivalent to $y2(n+1)−y2(n)≥r2l−a2u(M2+ε)−β2u(U2+ε).$(40)

Summing both sides of (40) from nτ1(n) to n − 1 leads to $∑j=n−τ1(n)n−1(y2(n+1)−y2(n))≥∑j=n−τ1(n)n−1[r2l−a2u(M2+ε)−β2u(U2+ε)]≥[r2l−a2u(M2+ε)−β2u(U2+ε)]τ~.$

Then $y2(n−τ1(n))≤y2(n)−[r2l−a2u(M2+ε)−β2u(U2+ε)]τ~.$

Thus $x2(n−τ1(n))≤x2(n)exp⁡{−[r2l−a2u(M2+ε)−β2u(U2+ε)]τ~}.$(41)

Substituting (33) and (41) into the second equation of (7), we have $x2(n+1)≥x2(n)exp⁡{r2l−β2u(U2+ε)−a2ux2(n)×exp⁡{−[r2l−a2u(M2+ε)−β2u(U2+ε)]τ~}}$(42) for all n > N3 + τ.

By applying Lemmas 2.2 and 2.4, we immediately get $limn→+∞infx2(n)≥m2ε,$(43) where $m2ε=r2l−β2u(U2+ε)a2uexp⁡{−[r2l−a2u(M2+ε)−β2u(U2+ε)]τ~}×exp⁡{r2l−β2u(U2+ε)−a2uexp⁡{−[r2l−a2u(M2+ε)−β2u(U2+ε)]τ~}M2}.$

Setting ε → 0 in (43), then $limn→+∞infx2(n)≥m2,$(44) where $m2=r2l−β2uU2a2uexp⁡{−[r2l−a2uM2−β2uU2]τ~}×exp⁡{r2l−β2uU2−a2uexp⁡{−[r2l−a2uM2−β2uU2]τ~}M2}.$

Without the loss of generality, we assume that $\begin{array}{}\epsilon <\frac{1}{2}min\left\{{m}_{1},{m}_{2}\right\}.\end{array}$ For any positive constant ε small enough, it follows from (38) and (44) that there exists large enough N4 >N3 + τ such that $x1(n)≥m1−ε,x2(n)≥m2−ε$(45) for any nN4.

From the third and fourth equations of system (7) and (45), we can derive that $Δu1(n)≥−γ1(n)u1(n)+η1(n)(m1−ε),$(46) $Δu2(n)≥−γ2(n)u2(n)+η2(n)(m2−ε).$(47)

Hence $u1(n+1)≥(1−γ1u)u1(n)+η1l(m1−ε),$(48) $u2(n+1)≥(1−γ2u)u2(n)+η2l(m2−ε).$(49)

By applying Lemmas 2.2 and 2.3, we immediately get $limn→+∞infu1(n)≥η1l(m1−ε)γ1u,$(50) $limn→+∞infu2(n)≥η2l(m2−ε)γ2u.$(51)

Setting ε → 0 in the above inequality leads to $limn→+∞infu1(n)≥η1lm1γ1u:=U1l,$(52) $limn→+∞infu2(n)≥η2lm2γ2u:=U2l.$(53)

This completes the proof of Theorem 2.6.

## 3 Existence and stability of periodic solution

In this section, we will study the stability of (7) under the assumption τi(n) = 0(i = 1,2), namely, we consider the following system $x1(n+1)=x1(n)exp⁡{r1(n)−a1(n)x1(n)−b1(n)(x2(n)−c2(n))2−β1(n)u1(n)},x2(n+1)=x2(n)exp⁡{r2(n)−a2(n)x2(n)+b2(n)(x1(n)−c1(n))2−β2(n)u2(n)},Δu1(n)=−γ1(n)u1(n)+η1(n)x1(n),Δu2(n)=−γ2(n)u2(n)+η2(n)x2(n).$(54)

Throughout this section we always assume that ri(n), ai(n), bi(n),ci(n), γi(n) and η1(n) are all bounded nonnegative periodic sequences with a common period ω and satisfy $0<γi(n)<1,n∈N∩[0,ω],i=1,2.$(55)

Also it is assumed that the initial conditions of (54) are of the form $xi(0)>0,ui(0)>0,i=1,2.$(56)

In similar way, we can derive the permanence of (54). We still let Mi and Ui be the upper bound of {xi(n)} and {ui(n)}, and mi and $\begin{array}{}{U}_{i}^{l}\end{array}$ be the lower bound of {xi(n)} and {ui(n)}.

#### Theorem 3.1

In addition to (55), assume that (H1) and (H2) $\begin{array}{}{r}_{1}^{l}>{b}_{1}^{u}\left({M}_{2}+{c}_{2}^{u}{\right)}^{2}+{\beta }_{1}{U}_{1},{r}_{2}^{l}>{\beta }_{2}^{u}{U}_{2}\end{array}$ hold, then system (54) has a periodic ω solution denoted by $\begin{array}{}\left\{{\overline{x}}_{1}\left(n\right),{\overline{x}}_{2}\left(n\right),{\overline{u}}_{1}\left(n\right),{\overline{u}}_{2}\left(n\right)\right\}.\end{array}$

#### Proof

Let $\begin{array}{}\mathrm{\Omega }=\left\{\left({x}_{1},{x}_{2},{u}_{1},{u}_{2}\right)|{m}_{i}\le {x}_{i}\le {M}_{j},{U}_{i}^{l}\le {u}_{i}\le {U}_{i},i=1,2\right\}.\end{array}$ It is easy to see that Ω is an invariant set of system (54). Then we can define a mapping F on Ω by $F(x1(0),x2(0),u1(0),u2(0))=(x1(ω),x2(ω),u1(ω),u2(ω))$(57) for (x1(0), x2(0), u1(0),u2(0)) ∈ Ω. Obviously, F depends continuously on (x1(0), x2(0), u1(0), u2(0)) . Thus F is continuous and maps a compact set Ω into itself. Therefore, F has a fixed point $\begin{array}{}\left({\overline{x}}_{1}\left(n\right),{\overline{x}}_{2}\left(n\right),{\overline{u}}_{1}\left(n\right),{\overline{u}}_{2}\left(n\right)\right).\end{array}$ So we can conclude that the solution $\begin{array}{}\left({\overline{x}}_{1}\left(n\right),{\overline{x}}_{2}\left(n\right),{\overline{u}}_{1}\left(n\right),{\overline{u}}_{2}\left(n\right)\right)\end{array}$ passing through $\begin{array}{}\left({\overline{x}}_{1},{\overline{x}}_{2},{\overline{u}}_{1},{\overline{u}}_{2}\right)\end{array}$ is a periodic solution of system (54). The proof of Theorem 3.1 is complete.

Next, we investigate the global stability property of the periodic solution obtained in Theorem 3.1.

#### Theorem 3.2

In addition to the conditions of Theorem 3.1, assume that the following condition (H3) hold, $H3χ1=max{|1−a1lm1|,|1−a1uM1|}+2b1uc2u[M22+M2]+β1u<1χ2=max{|1−a2lm2|,|1−a2uM2|}+2b2uc1u[M12+M1]+β2u+<1,χ3=(1−γ1l)+η1uM1<1,χ4=(1−γ2l)+η2uM2<1,$ then the ω periodic solution $\begin{array}{}\left({\overline{x}}_{1}\left(n\right),{\overline{x}}_{2}\left(n\right),{\overline{u}}_{1}\left(n\right),{\overline{u}}_{2}\left(n\right)\right)\end{array}$ obtained in Theorem 3.1 is globally attractive.

#### Proof

Assume that (x1(n), x2(n), u1(n), u2(n)) is any positive solution of system (54). Let $xi(n)=x¯i(n)exp⁡{yi(n)},ui(n)=u¯i(n)+vi(n),i=1,2.$(58)

To complete the proof, it suffices to show $limn→∞yi(n)=0,limn→∞vi(n)=0,i=1,2.$(59)

Since $y1(n+1)=y1(n)−a1(n)x¯1(n)[exp⁡(y1(n))−1]−b1(n)[(x2(n)−c2(n))2−(x¯2(n)−c2(n))2]−β1(n)v1(n)=y1(n)−a1(n)x¯1(n)exp⁡{θ1(n)y1(n)}y1(n)−b1(n)x¯22(n)exp⁡{2θ2(n)y2(n)}2y2(n)+2b1(n)c2(n)x¯2(n)exp⁡{θ3(n)y2(n)}y2(n)−β1(n)v1(n),$(60) where θi(n) ∈ (0,1), i = 1, 2, 3. In a similar way, we get $y2(n+1)=y2(n)−a2(n)x¯2(n)exp⁡{θ4(n)y2(n)}y2(n)+b2(n)x¯12(n)exp⁡{2θ5(n)y1(n)}2y1(n)+2b2(n)c1(n)x¯1(n)exp⁡{θ6(n)y1(n)}y1(n)−β2(n)v2(n),$(61) where θi(n) ∈ (0,1), j = 4, 5, 6.

Also, one has $v1(n+1)=(1−γ1(n))v1(n)+η1(n)x¯1(n)[exp⁡{y1(n)}−1]=(1−γ1(n))v1(n)+η1(n)x¯1(n)exp⁡{θ7(n)y1(n)}y1(n)$(62) $v2(n+1)=(1−γ2(n))v2(n)+η2(n)x¯2(n)[exp⁡{y2(n)}−1]=(1−γ2(n))v2(n)+η2(n)x¯2(n)exp⁡{θ8(n)y2(n)}y2(n).$(63)

By (H3), we can choose a ε > 0 such that $χ1ϵ=max{|1−a1l(m1−ϵ)|,|1−a1u(M1+ϵ)|}+2b1uc2u[(M2+ϵ)2+M2+ϵ]+β1u<1χ2ϵ=max{|1−a2l(m2−ϵ)|,|1−a2u(M2+ϵ)|}+2b2uc1u[(M1+ϵ)2+M1+ϵ]+β2u<1,χ3ϵ=(1−γ1l)+η1u(M1+ϵ)<1,χ4ϵ=(1−γ2l)+η2u(M2+ϵ)<1,$(64) In view of Proposition 2.5 and Theorem 2.6, there exists N5 > N4 such that $mi−ϵ≤xi(n),x¯i(n)≤Mi+ϵ,forn≤N5,i=1,2.$(65)

It follows from (60) and (61) that $y1(n+1)≤max{|1−a1l(m1−ϵ)|,|1−a1u(M1+ϵ)|}|y1(n)|+2b1uc2u[(M2+ϵ)2+M2+ϵ]|y2(n)|+β1u|v1(n)|,$(66) $y2(n+1)≤max{|1−a2l(m2−ϵ)|,|1−a2u(M2+ϵ)|}|y2(n)|+2b2uc1u[(M1+ϵ)2+M1+ϵ]|y1(n)|+β2u|v2(n)|,$(67) Also, for n > N5, one has $v1(n+1)≤(1−γ1l)|v1(n)|+η1u(M1+ϵ)|y1(n)|,$(68) $v2(n+1)≤(1−γ2l)|v2(n)|+η2u(M2+ϵ)|y2(n)|.$(69)

Let $\chi =max\left\{{\chi }_{1}^{ϵ},{\chi }_{2}^{ϵ},{\chi }_{3}^{ϵ},{\chi }_{4}^{ϵ}\right\},$ then 0 < χ < 1. It follows from (66) -(69) that $max{|y1(n+1)|,|y2(n+1)|,|v1(n+1)|,|v2(n+1)|}≤χmax{|y1(n)|,|y2(n)|,|v1(n)|,|v2(n)|}$(70)

for n > N5. Then we get $max{|y1(n)|,|y2(n)|,|v1(n)|,|v2(n)|}≤χn−N5max{|y1(N5)|,|y2(N5)|,|v1(N5)|,|v2(N5)|}.$(71)

Thus $limn→∞yi(n)=0,limn→∞vi(n)=0,i=1,2.$(72)

This completes the proof.

#### Remark 3.3

Although Zhi et al. [46] have investigated the permanence and almost periodic solution for an enterprise cluster model based on ecology theory with feedback controls on time scales, they also do not consider the enterprise cluster model with time delays, moveover, they do not investigate the periodic solution of this model. In this paper, we consider the permanence and periodic solutions of two enterprises with multiple delays, which is more general than those models in [46]. Thus our results complement the previous work of [46].

## 4 Examples

#### Example 4.1

Consider the following system $x1(n+1)=x1(n)exp⁡{r1(n)−a1(n)x1(n−τ1(n))−b1(n)(x2(n−τ2(n))−c2(n))2−β1(n)u1(n)},x2(n+1)=x2(n)exp⁡{r2(n)−a2(n)x2(n−τ1(n))+b2(n)(x1(n−τ2(n))−c1(n))2−β2(n)u2(n)},Δu1(n)=−γ1(n)u1(n)+η1(n)x1(n),Δu2(n)=−γ2(n)u2(n)+η2(n)x2(n),$(73) where r1(n) = 0.4 + sin(n), r2(n) = 0.5 + cos(n), a1(n) = a2(n) = 1, b1(n) = 0.4 + cos(n), b2(n) = 0.3 + cos(n), c1(n) = 0.4 + cos(n), c2(n) = 0.4 + cos(n), β1(n) = 0.2 + cos(n), β2(n) = 0.4 + sin(n), γ1(n) = cos(n) + 0.3, γ2(n) = sin(n) + 0.4, η1(n) = 0.5 + cos(n), η2(n) = 0.7 + cos(n), τi(n) = 1. We can verify that all the assumptions in Theorem 2.6 are fulfilled. Then (73) is permanent which is shown in Figures 1-2.

Fig. 1

Times series of x1 and x2 for system (73), where the blue line stands for x1 and the red line stands for x2.

Fig. 2

Times series of u1 and u2 for system (73), where the blue line stands for u1 and the red line stands for u2.

#### Example 4.2

Consider the following system $x1(n+1)=x1(n)exp⁡{r1(n)−a1(n)x1(n)−b1(n)(x2(n)−c2(n))2−β1(n)u1(n)},x2(n+1)=x2(n)exp⁡{r2(n)−a2(n)x2(n)+b2(n)(x1(n)−c1(n))2−β2(n)u2(n)},Δu1(n)=−γ1(n)u1(n)+η1(n)x1(n),Δu2(n)=−γ2(n)u2(n)+η2(n)x2(n),$(74) where r1(n) = 0.4 + sin(n), r2(n) = 0.5 + sin(n), a1(n) = a2(n) = 1, b1(n) = 0.4 − cos(n), b2(n) = 0.3 − cos(n), c1(n) = 0.4 + sin(n), c2(n) = 0.3 + sin(n), α1(n) = 0.3 + cos(n), β2(n) = 0.2 + sin(n), γ1(n) = cos(n) + 0.2, γ2(n) = sin(n) + 0.3, η1(n) = 0.2 + cos(n), η2(n) = 0.2 + cos(n). We can verify that all the assumptions in Theorem 3.1 are fulfilled. Then we know that the periodic solution of system (74) is globally attractive which is illustrated in Figures 3-3.

Fig. 3

Times series of x1 and x2 for system (74), where the blue line stands for x1 and the red line stands for x2.

Fig. 4

Times series of u1 and u2 for system (74), where the blue line stands for u1 and the red line stands for u2.

## 5 Conclusions

In the present article, a discrete system with competition and cooperation model of two enterprises is proposed. Applying the difference inequality theory, we have established some sufficient conditions which guarantee the permanence of the system. We find that under some suitable conditions, the competition and cooperation of enterprises cluster can remain balanced. This shows that feedback control effect and time delays play a key role in deciding the survival of enterprises. The sufficient conditions which ensure the existence and stability of unique globally attractive periodic solution of the system without time delays are established. The obtained results are completely new and complement the published works of [2529, 46].

## Acknowledgements

The first and second authors were supported by the Key Research Institute of Philosophies and Social Sciences in Guangxi Universities and Colleges(16YC001, 16YC002). The third author was supported by Key Research Institute of Philosophies and Social Sciences in Guangxi Universities and Colleges(16YC001, 16YC002) and Key Project of Science and Technology Research in Guangxi Universities and Colleges(ZD2014058). The authors would like to thank the referees and the editor for helpful suggestions incorporated into this paper.

## References

• [1]

Wang X.H., Huang C.Y., Permanence of a stage-structured predator-prey system with impulsive stocking prey and harveting predator, Appl. Math. Comput., 2014, 235, 32-42

• [2]

Mukherjee D., Permanence and global attractivity for facultative mutualism system with delay, Math. Methods Appl. Sci., 2003, 26, 1-9

• [3]

Zhao J.D., Jiang J.F., Average conditons for permanence and extinction in nonantonomous Lotka-Volterra system, J. Math. Anal. Appl., 2004, 299, 663-675

• [4]

Teng Z.D., Zhang Y., Gao S.J., Permanence criteria for general delayed discrete nonautonomous -species Kolmogorov systems and its applications, Comput. Math. Appl., 2010, 59, 812-828

• [5]

Liu Z.J., Zhong S.M., Liu X.Y., Permanence and periodic solutions for an impulsive reaction-diffusion food-chain system with ratiodependent functional response, Commun. Nonlinear Sci. Numer. Simul., 2014, 19, 173-188

• [6]

Chen F.D., Permanence of a discrete n-species food-chain system with time delays, Appl. Math. Comput., 2007, 185, 719-726

• [7]

Dhar J., Jatav K.S., Mathematical analysis of a delayed stage-structured predator-prey model with impulsive diffusion between two predators territories, Ecol. Complex., 2013, 16, 59-67

• [8]

Liu S.Q., Chen L.S., Necessary-sufficient conditions for permanence and extinction in lotka-volterra system with distributed delay, Appl. Math. Lett., 2003, 16, 911-917

• [9]

Liao X.Y., Zhou S.F., Chen Y.M., Permanence and global stability in a discrete n-species competition system with feedback controls, Nonlinear Anal.: Real World Appl., 2008, 9, 1661-1671

• [10]

Hu H.X., Teng Z.D., Jiang H.J., On the permanence in non-autonomous Lotka-Volterra competitive system with pure-delays and feedback controls, Nonlinear Anal.: Real World Appl., 2009, 10, 1803-1815

• [11]

Muroya Y., Permanence and global stability in a Lotka-Volterra predator-prey system with delays, Appl. Math. Lett., 2003, 16, 1245-1250

• [12]

Hou Z.Y., On permanence of Lotka-Volterra systems with delays and variable intrinsic growth rates, Nonlinear Anal.: Real World Appl., 2013, 14, 960-975

• [13]

Xiong X.S., Zhang Z.Q., Periodic solutions of a discrete two-species competitive model with stage structure, Math. Comput. Modelling, 2008, 48, 333-343

• [14]

Zhang R.Y., Wang Z.C., Chen Y.M., Wu J.H., Periodic solutions of a single species discrete population model with periodic harvest/stock, Comput. Math. Appl., 2009, 39, 77-90.Google Scholar

• [15]

Takeuchi Y., Global Dynamical Properties of Lotla-Volterra Systems, World Scientific, River Edge, NJ, USA, 1996Google Scholar

• [16]

Sun Y.G., Saker S.H., Positive periodic solutions of discrete three-level food-chain model of Holling type II, Appl. Math. Comput., 2006, 180, 353-365Google Scholar

• [17]

Ding X.H., Liu C., Existence of positive periodic solution for ratio-dependent N-species difference system, Appl. Math. Modelling, 2009, 33, 2748-2756

• [18]

Li Z.C., Zhao Q.L., Ling D., Chaos in a discrete population model, Discrete Dyn. Nat. Soc., 2012, 2012, Ariticle ID 482459, 14 pages

• [19]

Xiang H., Yang K.M., Wang B.Y., Existence and global stability of periodic solution for delayed discrete high-order hopfifled-type neural networks, Discrete Dyn. Nat. Soc., 2005, 2005, 281-297

• [20]

Lu C., Zhang L.J., Permanence and global attractivity of a discrete semi-ratio dependent predator-prey system with Holling II type functional response, J. Appl. Math. Comput., 2010, 33, 125-135

• [21]

Xu J.B., Teng Z.D., Jiang H.J., Permanence and global attractivity for discrete nonautonomous two-species Lotka-Volterra competitive system with delays and feedback controls, Period Math Hung., 2011, 63, 19-45

• [22]

Chen F.D., You M.S., Permanence for an integrodifferential model of mutualism, Appl. Math. Comput., 2007, 186, 30-34

• [23]

Chen F.D., Permanence in a discrete Lotka-Volterra competition model with deviating arguments, Nonlinear Anal.: Real World Appl., 2008, 9, 2150-2155

• [24]

Tian X.H., Nie Q.K., On model construction of enterprises’ interactive relationship from the perspective of business ecosystem, South. Econ. J., 2006, 4, 50-57 Google Scholar

• [25]

Liao M.X., Xu C.J., Tang X.H., Dynamical behavior for a competition and cooperation model of enterpries with two delays, Nonlinear Dyn., 2014, 75, 257-266

• [26]

Liao M.X., Xu C.J., Tang X.H., Stability and Hopf bifurcation for a competition and cooperation model of two enterprises with delay, Commun. Nonlinear Sci. Numer. Simul., 2014, 19, 3845-3856

• [27]

Li Y.K., Zhang T.W., Global asymptotical stability of a unique almost periodic solution for enterprise clusters based on ecology theory with time-varying delays and feedback controls, Commun. Nonlinear Sci. Numer. Simul., 2012, 17, 904-913

• [28]

Xu C.J., Periodic behavior of competition and corporation dynamical model of two enterprises on time scales, J. Quant. Econ., 2012, 29, 1-4 Google Scholar

• [29]

Xu C.J., Shao Y.F., Existence and global attractivity of periodic solution for enterprise clusters based on ecology theory with impulse, J. Appl. Math. Comput., 2012, 39, 367-384

• [30]

Agarwal R.P., Difference Equations and Inequalities: Theory, Method and Applications, vol. 228 of Monographs and Textbooks in Pure and Applied Mathematics, Marcel Dekker, New York, NY, USA, 2nd edition, 2000 Google Scholar

• [31]

Li Y.K., Lu L.H., Positive periodic solutions of discrete n-species food-chain systems, Appl. Math. Comput., 2005, 167, 324-344

• [32]

Chen X., Chen F.D., Stable periodic solution of a discrete periodic Lotka-Volterra competition system with a feedback control, Appl. Math. Comput., 2006, 181, 1446-1454Google Scholar

• [33]

Muroya Y., Persistence and global stability in Lotka-Volterra delay differential systems, Appl. Math. Lett., 2004, 17, 795-800

• [34]

Muroya Y., Partial survival and extinction of species in discrete nonautonomous Lotka-Volterra systems, Tokyo J. Math., 2005, 28, 189-200

• [35]

Yang X.T., Uniform persistence and periodic solutions for a discrete predator-prey system with delays, J. Math. Anal. Appl., 2006, 316, 161-177

• [36]

Chen F.D., Permanence for the discrete mutualism model with time delays, Math. Comput. Modelling, 2008, 47, 431-435

• [37]

Zhang W.P., Zhu D.M., Bi P., Multiple periodic positive solutions of a delayed discrete predator-prey system with type IV functional responses, Appl. Math. Lett., 2007, 20, 1031-1038

• [38]

Chen Y.M., Zhou Z., Stable periodic of a discrete periodic Lotka-Volterra competition system, J. Math. Anal. Appl., 2003, 277, 358-366

• [39]

Li X.P., Yang W.S., Permanence of a discrete model of mutualism with infinite deviating arguments, Discrete Dyn. Nat. Soc., 2010, 2010, Article ID 931798, 7 pages

• [40]

Li X.P., Yang W.S., Permanence of a discrete predator-prey systems with Beddington-DeAngelis functional response and feedback controls, Discrete Dyn. Nat. Soc., 2008, 2008, Article ID 149267, 8 pages

• [41]

Chen F.D., Permanence and global attractivity of a discrete multispecies Lotka-Volterra competition predator-prey systems, Appl. Math. Comput., 2006, 182, 3-12 Google Scholar

• [42]

Zhou H., Enterprises cluster co-existence model and stability analysis, Syst. Eng., 2003, 21, 32-37 Google Scholar

• [43]

Guo Q., Competitive strages in an enterprises: an ecological model, Contem. Econ. Manage, 2005, 27, 49-52Google Scholar

• [44]

Zhang R., Qian X.S., Gao Z.A., A competitive model of enterprises based on ecology theory, Syst. Eng., 2008, 26, 116-119 Google Scholar

• [45]

Wang L., Wang M.Q., Ordinary Difference Equation, Xinjiang University Press, China, 1991 Google Scholar

• [46]

Zhi Y.H., Ding Z.L., Li Y.K., Permanence and almost periodic solution for an enterprise cluster model based on ecology theory with feedback controls on time scales, Discrete Dyn. Nat. Soc., 2013, 2013, Article ID 639138, 14 pages

Accepted: 2017-01-26

Published Online: 2017-03-13

Citation Information: Open Mathematics, Volume 15, Issue 1, Pages 218–232, ISSN (Online) 2391-5455,

Export Citation