810 IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING, VOL. 11, NO.

3, JULY 2014

Reaching Law Approach to the Sliding Mode Control

of Periodic Review Inventory Systems
Andrzej Bartoszewicz and Piotr Leśniewski

Abstract—In this paper, a discrete-time sliding mode inven- sliding motion on the hypersurface, or applying the reaching law
tory management strategy based on a novel non-switching type approach. In the latter case the desirable evolution of the sliding
reaching law is introduced. The proposed reaching law eliminates variable is first specified, and then a controller which ensures
undesirable chattering, and ensures that the sliding variable rate
of change is upper bounded by a design parameter which does not
that the variable changes according to the specification is deter-
depend on the system initial conditions. This approach guarantees mined. The reaching law approach was first introduced by Gao
fast convergence with non-negative, upper limited supply orders, and Hung for continuous time systems [18]. In that paper, con-
and ensures that the maximum stock level may be specified a priori stant, constant plus proportional, and power rate reaching laws
by the system designer. Furthermore, a sufficient condition for were considered. Then, in paper [19] (see also [4] for further
100% customers’ demand satisfaction is derived. The inventory comments), the idea of constant plus proportional rate reaching
replenishment system considered in this paper involves multiple
suppliers with different lead times and different transportation law has been extended to discrete-time systems. Since then the
losses in the delivery channels. reaching law approach to the control of discrete-time systems
has been used by many researchers [9], [20], [22], [28], [31],
Note to Practitioners—This paper presents a new periodic review [33], [34], [38]. Even though much research in this field has
inventory management strategy which prevents from exceeding already been done, the original approach proposed in [19] is
the available storage capacity, ensures smooth order evolution
still very popular. Therefore, in this paper, we extend the results
and helps attenuate the bullwhip effect. The strategy is scalable,
computationally efficient, and easy to implement in any typical of [19] in order to obtain a non-switching discrete-time sliding
inventory replenishment system. The strategy explicitly accounts mode controller [3], [5] and to ensure faster convergence of the
for transportation losses and different lead times of commodity controlled system without increasing the magnitude of the con-
suppliers. trol signal. The first of the two objectives is accomplished with
Index Terms—Digital control, sliding mode control. the application of the quasi-sliding mode definition proposed in
[5], and the latter one is achieved by the introduction of a vari-
able, state dependent convergence rate factor in the proposed
I. INTRODUCTION reaching law. In the second part of the paper, we apply the pro-

C ONTINUOUS time variable structure and sliding mode posed reaching law to design a new periodic review inventory
control systems were originally introduced about 60 years replenishment strategy [8], [21], [22], [27], [36], [37], [39] for a
ago in Russia [15], [40]. Their exceptional robustness [13] and warehouse with multiple remote suppliers and delivery channels
good computational efficiency, have immediately gained them characterized by different commodity loss factors. We demon-
much interest and many advocates in the control engineering strate favorable properties of the designed strategy which could
community [12], [14], [18]. A few years later, discrete-time not be achieved with the application of the original constant
sliding mode control systems have also been proposed [32], [41] plus proportional reaching law. In particular, we show that our
and then analyzed in numerous significant studies [1]–[3], [5], reaching law ensures non-negative upper bounded supply or-
[10], [11], [16], [17], [19], [20], [23]–[26], [28]–[31], [33], [35], ders which do not depend on the warehouse capacity, and there-
[42]–[44]. fore are fairly desirable in the considered system. Furthermore,
Both discrete and continuous time sliding mode controllers we demonstrate that our reaching law-based controller elimi-
push the system representative point (state vector) onto a prede- nates the risk of exceeding warehouse capacity and may ensure
termined hypersurface in the state space. This can be achieved in full customers’ demand satisfaction. The work presented in this
two different ways. Either assuming a certain control algorithm paper differs from our earlier results [22] in the following three
and demonstrating that this algorithm guarantees stability of the aspects. First, a totally new reaching law appropriate for any dy-
namic system is proposed in this paper, second, the supply chain
model considered here explicitly takes into account transporta-
Manuscript received November 26, 2013; revised February 03, 2014; ac-
cepted March 25, 2014. Date of publication April 17, 2014; date of current ver- tion losses which was not the case in [22], and finally a novel
sion June 30, 2014. This paper was recommended for publication by Associate feasible order allocation among various remote providers is in-
Editor A.-S. Jia and Editor L. Shi upon evaluation of the reviewers’ comments. troduced.
This work was supported in part by the National Science Centre of Poland
under decision number DEC 2011/01/B/ST7/02582 under the framework of
project “Optimal Sliding Mode Control of Time Delay Systems” and in part
by the Foundation for Polish Science under the Mistrz grant.
II. NON-SWITCHING REACHING LAW
The authors are with the Institute of Automatic Control, Technical University
of Łódź, 90-924 Łódź, Poland (e-mail: [email protected]; piotr.
Let us consider the following discrete-time system:
lesniewski2@ gmail.com).
Digital Object Identifier 10.1109/TASE.2014.2314690 (1)

1545-5955 © 2014 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/
redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
BARTOSZEWICZ AND LEŚNIEWSKI: REACHING LAW APPROACH TO THE SLIDING MODE CONTROL OF PERIODIC REVIEW INVENTORY SYSTEMS 811

where is the state vector , is the state generated in the system, and sluggish convergence to the
matrix, is the model uncertainty matrix, is the input vector, vicinity of . The proposed reaching law has two
is a scalar input, and is a disturbance vector. We major advantages over the one presented in [19]. First, it
denote the demand state vector by , and define the closed-loop does not contain a discontinuous term, so it does not lead to
system error as . Then, we select the sliding chattering. Second, since increases with the decrease
variable as of , our reaching law results in faster convergence and
better robustness with the same bounds on the control signal
(2) magnitude.
In order to find the control signal which ensures that
With this choice of variable , equation deter- the sliding variable evolution is indeed described by (3), we use
mines the sliding hyperplane. The elements of (1) to rewrite (2) as follows:
vector are selected in such a way that and that the
closed-loop system exhibits the desired performance. This can
be done in a few ways including quadratic optimization [26], (10)
pole placement method [19], dead-beat design [6], [23], etc.
In this paper, the quasi-sliding mode is defined similarly as in Then, comparing (3) and (10), we obtain
[5], i.e., it is such a motion of the system that its representative
point (state) remains in a given band around sliding hyperplane
, where is defined by (2). According to this (11)
definition, the representative point (state of the system) in the
quasi-sliding mode is confined to a specified vicinity of the hy- Since all terms in (11) are either constants, or variables which do
perplane. Contrary to the definition introduced in [19], in this not depend on unknown terms or , this is a feasible
paper, crossing the hyperplane is allowed but not required. control signal which can actually be implemented in the system
Let us now consider the following reaching law considered in this paper.
In the next two theorems, we demonstrate that once the rep-
resentative point of system (1) has reached a band around the
sliding hyperplane , it remains inside the band, and
(3)
also that the proposed reaching law makes the point always
where move towards this band.
Theorem 1: If the following inequality:
(4)
(12)
represents the influence of the model uncertainty on the sliding
variable evolution and is satisfied at some instant , then it is also true for any
.
(5) Proof: From (3), we observe that increases
with the increase of . Therefore, even assuming the
denotes the effect of disturbance on this variable. Furthermore,
most disadvantageous possible influence of the disturbance and
and are the mean values of and , namely
model uncertainty, if (12) is satisfied for some , then from (3),
we obtain
(6)

where are upper and lower bounds of , and are

upper and lower bounds of , i.e.,

(7) (13)
The notation used in (4)–(7) is adopted from [19].
Convergence rate factor in (3) is given by Using this observation and assumption (12) by virtue of the prin-
ciple of mathematical induction, we conclude that (12) indeed
(8) holds for all .
Theorem 2: If the absolute value of is greater than the
where is a design constant. The constant is chosen so that right hand side of (12), then converges, at least asymp-
, where and represent the greatest possible totically, to the band specified by (12).
deviation of and from their mean values Proof: In the proof, we will consider two cases, namely,
the positive and negative values of .
(9) Case 1: If

Appropriate choice of allows to find a satisfactory com-

promise between excessive magnitudes of the control signal (14)
812 IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING, VOL. 11, NO. 3, JULY 2014

Fig. 1. Inventory supply model.

then using (3), we obtain commodities are generated by the controller located at the distri-
bution center. The control signal determines the total amount
of supplies requested from all of the providers. This value is dis-
tributed among the providers, proportionally to the maximum
amount of goods they can send, i.e., supplier receives an order
equal to . The block diagram of the
periodic review inventory system considered in this section is
shown in Fig. 1. It is assumed, that each lead time is a mul-
tiple of the review period , i.e., , where is a posi-
tive integer. In fact, this is a well justified assumption, since even
if the actual order procurement time is a non-integer multiple of
(15)
the review period, still the arrival of goods at the warehouse
is detected by the enterprise management system only at dis-
Since , the difference crete-time instants. Therefore, this time is actually rounded up to
is negative and it approaches zero only if tends the nearest integer multiple of . The warehouse stock level at
to zero. This proves that if initial value of is positive, then it time is denoted by . The consumers’ demand is mod-
asymptotically converges to the band specified by relation (12). eled by an a priori unknown function of time , bounded
Case 2: Similarly if by a known constant . If the amount of stored goods is in-
sufficient, the demand cannot be fully covered. Therefore, an
additional function is introduced, which represents the
amount of goods actually sold to the customers. For any ,
(16)
the following inequalities hold:
then again, using (3), we obtain
(18)

(17) The warehouse is assumed to be empty prior to the beginning

Inequality (17) shows, that if (16) is satisfied, then difference of the control process, i.e., . Moreover, the first
is positive and it approaches zero only if order is sent at , i.e., . The inventory
tends to zero. stock level for can therefore be obtained as the differ-
Taking into account the conclusions of both cases, we find ence between all acquired and sold goods
that if lies outside the band around specified
by (12), then it asymptotically converges to this band. (19)

III. INVENTORY SUPPLY MODEL

where .
In this section, we consider an inventory management system
We can represent all providers with equal lead times as a
with remote providers. The transportation channel between
single “aggregate” supplier, so as to get a simplified, equivalent
the th provider and the warehouse is character-
version of the system model. The amount of commodities that
ized by its lead time and commodity loss factor .
arrive at the distribution center from this “aggregate” supplier
Furthermore, each of the providers has its own maximum ad-
is equal to , where
missible supply rate, i.e., the greatest amount of goods that it
can send during one review period. This amount, for the th
provider is denoted by .
(20)
The commodities obtained from the providers are used to sat-
isfy an a priori unknown consumers’ demand. The orders for the
BARTOSZEWICZ AND LEŚNIEWSKI: REACHING LAW APPROACH TO THE SLIDING MODE CONTROL OF PERIODIC REVIEW INVENTORY SYSTEMS 813

for and . Naturally, if no and substitute it into (22). This results in the following state
provider has the lead time , then the corresponding coefficient matrix of the closed-loop system:
. Now, we can express the stock level as follows:

(21)
.. .. ..
. . .
We can also represent the above relation in the standard state
space form
(27)

(22) which has the following characteristic polynomial

where is the state

vector, is the on-hand stock level. The re-
maining state variables are the delayed values of the control
signal, i.e., (28)

(23)
We have already assumed that . This condition and rela-
for . is state matrix tion (25), imply . A linear discrete-time system is asymp-
totically stable if and only if all of its eigenvalues lie inside a unit
circle on the -plane. Moreover, to obtain finite time error con-
.. .. vergence to zero the characteristic polynomial (28) should have
..
. . . (24) the following form:

(29)

and , , and are vectors We find that (28) reduces to (29) when vector is chosen as
follows:
..
. .. .. (25)
. . (30)
for

The desired state of the system is , where

is the demand level of the on-hand stock. In general, the We observe, that with this choice of vector the variable
bigger parameter is chosen, the greater amount of goods has has a clear physical meaning, i.e., it represents the difference be-
to be accommodated in the warehouse and the better demand tween the desired warehouse stock level and the sum of amounts
satisfaction is achieved. Still, the effect of this parameter on the of goods in the warehouse and currently in transit.
overall system performance will be more precisely analyzed in We will now implement the proposed reaching law to derive
the next section. the controller that steers the representative point of the system
Closer analysis of the system described in this paper reveals to the proximity of the sliding hyperplane , where
that its total control signal is limited by the following constraint: is given by (30). For the considered periodic review inventory
. This precludes the use of a linear system, the perturbations of the sliding variable caused by the
controller, as the magnitude of its output would strongly depend model uncertainty and disturbance are
on the initial conditions of the system and could not satisfy the
constraint. Therefore, in the next section, we propose a non- (31)
linear sliding mode controller.
Using (2) and (31) with (11), we obtain
IV. CONTROLLER DESIGN
In this section, we will develop a sliding mode controller that
ensures the desired sliding variable evolution. We begin by se-
lecting the elements of vector , which describe the sliding hy-
perplane, so that the closed-loop system exhibits dead-beat char-
acteristics. First, we calculate the control signal needed to sat- (32)
isfy as
One can easily notice, that by selecting according to (30), we
(26) have obtained . Therefore, using (25) and
814 IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING, VOL. 11, NO. 3, JULY 2014

(30), we can obtain the control signal, which ensures the desired bution center. In the following theorem, we will determine the
sliding variable evolution as upper bound of the on-hand stock. Therefore, if warehouse ca-
pacity equal to or greater than this bound is secured, then there
will be no risk of hiring (usually very costly) emergency storage
(33) space.
Theorem 4: With the application of the proposed control
strategy, for every , the inventory stock level will satisfy
Let us notice at this point, that any other choice of vector the following condition:
would lead to a more convoluted expression determining
and less computationally efficient controller. Moreover, it is (37)
worth to point out that application of other hyperplane design
methods (pole placement or quadratic optimization) in conjunc- Proof: From (36), we obtain
tion with the reaching law approach is redundant as both the
reaching law approach and these methods are used primarily (38)
to satisfy input and state constraints of the controlled systems.
These observations justify the choice of the sliding hyperplane for any . Using (2) and (23), we can rewrite (38) as
determined by (30).
In the remainder of this section, important properties of the
proposed control strategy will be stated in three theorems and
proved. In the first one, we will demonstrate, that control signal (39)
(33) is always non-negative and upper bounded by an a priori We have already proven that the control signal is always non-
known constant. Since this signal directly corresponds to the negative. Therefore, we conclude that (39) implies (37).
amounts of goods sent by the providers, both of these features In order to obtain the greatest possible profit, one may wish to
are essential for the practical application of the proposed eliminate lost sales risk. Therefore, it is reasonable to establish
strategy. conditions ensuring that the consumers’ demand is always fully
Theorem 3: For any , control signal (33) satisfies the satisfied. For that purpose, in the last theorem, we determine
following two inequalities: the smallest value of the demand inventory stock level ensuring
that after some initial time, the warehouse will not be empty.
We can notice from (21) that this implies full satisfaction of the
(34)
customers’ demand.
Theorem 5: If the following condition is satisfied:

Proof: As shown in Theorems 1 and 2, sliding variable

will start at some initial value , and in each consecu-
(40)
tive step its absolute value will decrease unless (12) is satisfied.
Moreover, once (12) becomes true, it will hold for the rest of the
control process. For the system under consideration
then for any , where is the first
(35) time instant when (12) is true.
Proof: Using (12), for any , we obtain
Using (12), (31), and (35), we observe that

(36) (41)

for all . Moreover, substituting (12) into (33), we get

We now notice, that the value of the control signal (33) is al-
ways increasing with the increase of . Therefore, its min- (42)
imum value will be generated for the smallest possible ,
and the maximum value for the greatest . This observa-
tion allows us to simply substitute the limits of interval (36) into
which is also true for any . By combining (41) and (42),
(33) and conclude that (34) indeed holds.
we arrive at
We have assumed that the requests for goods are distributed
among the providers according to the maximum amount of
goods that they can deliver. Therefore, if is selected in such
a way that the right-hand side of (34) is equal to , (43)
then no supplier will be requested to send more goods, than it
is actually able to provide.
Any successful inventory management strategy should guar- for any . Therefore, if (40) holds, then the right-
antee that all of the incoming goods can be stored in the distri- hand side of the above inequality is always strictly positive.
BARTOSZEWICZ AND LEŚNIEWSKI: REACHING LAW APPROACH TO THE SLIDING MODE CONTROL OF PERIODIC REVIEW INVENTORY SYSTEMS 815

TABLE I
SIMULATION PARAMETERS

Fig. 4. On-hand stock level.

Fig. 2. Consumers’ demand.

Fig. 5. Sliding variable evolution.

than or equal to . Therefore, no sup-

plier is at any time expected to provide more commodities than
it is really able to. Furthermore, comparing Figs. 2 and 3, we
Fig. 3. Control signal. notice that the supply orders are significantly smoother than
the consumers’ demand is. This shows that our strategy attenu-
ates the highly undesirable bullwhip effect. The inventory stock
V. SIMULATION RESULTS level is illustrated in Fig. 4. As stated in Theorems 4 and 5 it
In order to verify the properties of the proposed control law never exceeds 594 items, and after some initial time it does not
computer simulations of an inventory system with three sup- drop to zero any more. This means, that the risk of hiring costly
pliers were performed. The review period day and the pa- emergency storage is eliminated, and full consumers’ demand
rameters of the suppliers and transportation channels are shown satisfaction is ensured. The evolution of the sliding variable is
in Table I. The greatest lead time equals 11 days, which implies shown in Fig. 5. As determined by Theorems 1 and 2 the vari-
, and . The elements of the first row able converges to the band (the band limits are
of matrix are , , , and the shown by dashed lines), and after reaching the band, the vari-
remaining elements are equal to zero. The maximum daily able does not leave it for the rest of the control process.
consumers’ demand . The actual evolution of We also consider the performance of the system when all its
the demand in the simulation example is shown in Fig. 2. The parameters, and the consumers’ demand transient are the same
demand exhibits abrupt changes between small and large values, as in the first scenario, but initially the warehouse is not empty.
which reflect the most adverse conditions that can appear in the We analyze two cases: the first one when the initial stock level
considered system. The total amount of goods, which can be is equal to the half of its maximum value,
provided by all suppliers on a single day is and the second one when the level equals
. Therefore, in order to ensure that the control signal its maximum value. Since the initial on-hand stock affects only
never exceeds this value, we select . As–ac- the beginning of the control process, in Figs. 6– 8, we depict the
cording to Theorem 5–the minimum demand value of the stock transients until the 25th day. This is justified by the fact, that
level that ensures full consumers’ demand satisfaction is 513 later on the transients are almost identical to each other and also
items, we select equal to 530 items. The simulation results to those shown in Figs. 3–5.
are presented in Figs. 3– 5. The control signal is shown in Fig. 3. As we can observe from Figs. 6–8 all of the advantageous
One can easily see that it is always non-negative and smaller properties of the proposed controller mentioned before, hold
816 IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING, VOL. 11, NO. 3, JULY 2014

inventory systems with multiple suppliers and transportation

losses. The reaching law introduced in this paper, as opposed
to the ones previously proposed in literature does not require
crossing the sliding hyperplane in each successive control step
of the quasi-sliding motion. This property eliminates chattering
and allows fast convergence with limited control signal. The
proposed reaching law based inventory management strategy
ensures non-negative and upper bounded supply orders, elim-
inates the risk of costly emergency storage, and may ensure that
no business opportunities are lost. These important properties
have been proved analytically and verified by computer simula-
tions. Our further research efforts will be focused on inventory
Fig. 6. Control signal for: (a) and (b) systems with different customer classes, i.e., customers whose
.
orders can be backordered and those whose orders must be re-
jected if they cannot be satisfied immediately from on-hand in-
ventory [7]. We will also employ the reaching law proposed in
this paper for other applications. This is possible as our reaching
law is not customized, but may be directly used with all other
types of hyperplanes or even nonlinear hypersufaces.

REFERENCES
[1] B. Bandyopadhyay and D. Fulwani, “High-performance tracking con-
troller for discrete plant using nonlinear sliding surface,” IEEE Trans.
Ind. Electron., vol. 56, pp. 3628–3637, Sep. 2009.
[2] B. Bandyopadhyay and S. Janardhanan, Discrete-Time Sliding Mode
Control. A Multirate Output Feedback Approach. Berlin, Germany:
Springer-Verlag, 2006.
[3] G. Bartolini, A. Ferrara, and V. Utkin, “Adaptive sliding mode control
Fig. 7. On-hand stock level for: (a) and (b) in discrete-time systems,” Automatica, vol. 31, pp. 769–773, 1995.
. [4] A. Bartoszewicz, “Remarks on discrete-time variable structure control
systems,” IEEE Trans. Ind. Electron., vol. 43, pp. 235–238, 1996.
[5] A. Bartoszewicz, “Discrete time quasi-sliding mode control strate-
gies,” IEEE Trans. Ind. Electron., vol. 45, no. 4, pp. 633–637, Aug.
1998.
[6] A. Bartoszewicz and J. Żuk, “Discrete-time sliding mode flow
controller for multi-source connection-oriented communication net-
works,” J. Vibration Control, vol. 15, no. 11, pp. 1745–1760, 2009.
[7] S. Benjaafar and M. Elhafsi, “A production-inventory system with both
patient and impatient demand classes,” IEEE Trans. Autom. Sci. Eng.,
vol. 9, no. 1, pp. 148–159, Jan. 2012.
[8] M. Boccadoro, F. Martinelli, and P. Valigi, “Supply chain management
by H-infinity control,” IEEE Trans. Autom. Sci. Eng., vol. 5, no. 4, pp.
703–707, Oct. 2008.
[9] S. Chakrabarty and B. Bandyopadhyay, “Quasi sliding mode control
with quantization in state measurement,” in Proc. IEEE 37th Annu.
Conf. Ind. Electron. Soc., 2011, pp. 3971–3976.
[10] M. L. Corradini and G. Orlando, “Variable structure control of dis-
cretized continuous-time systems,” IEEE Trans. Autom. Control, vol.
Fig. 8. Sliding variable evolution for: (a) and (b) 43, no. 9, pp. 1329–1334, Sep. 1998.
. [11] M. L. Corradini, V. Fossi, A. Giantomassi, G. Ippoliti, S. Longhi, and
G. Orlando, “Discrete time sliding mode control of robotic manipula-
tors: Development and experimental validation,” Control Eng. Prac-
tice, vol. 20, no. 8, pp. 816–822, 2012.
also for nonzero initial stock values. Comparing Fig. 3 and [12] R. S. DeCarlo, S. Żak, and G. Mathews, “Variable structure control of
Fig. 6(a), we can notice that the initial value of the control nonlinear multivariable systems: A tutorial,” Proc. IEEE, vol. 76, pp.
signal only marginally depends on the initial conditions, in 212–232, 1988.
[13] B. Draženović, “The invariance conditions in variable structure sys-
fact much less as compared to a linear controller. Furthermore, tems,” Automatica, vol. 5, pp. 287–295, 1969.
we can observe from Fig. 7 that if the initial stock level is [14] C. Edwards and S. Spurgeon, Sliding Mode Control: Theory and Ap-
sufficiently high, then it will not drop to zero, even before the plications. London, U.K.: Taylor & Francis, 1998.
[15] S. V. Emelyanov, Variable Structure Control Systems (in Russian).
start of the quasi-sliding motion, and full consumer demand Moscow, Russia: Nauka, 1967.
satisfaction will be ensured for any . [16] K. Furuta, “Sliding mode control of a discrete system,” Syst. Control
Lett., vol. 14, no. 2, pp. 145–152, 1990.
[17] Z. Galias and X. Yu, “Analysis of zero-order holder discretization of
VI. CONCLUSION two-dimensional sliding mode control systems,” IEEE Trans. Circuits
In this work, a new reaching law for discrete-time sliding Syst. II, vol. 55, no. 12, pp. 1269–1273, Dec. 2008.
[18] W. Gao and J. Hung, “Variable structure control of nonlinear systems:
mode control of dynamic plants has been introduced and applied A new approach,” IEEE Trans. Ind. Electron., vol. 40, no. 1, pp. 45–55,
to design a feasible management strategy for periodic review Feb. 1993.
BARTOSZEWICZ AND LEŚNIEWSKI: REACHING LAW APPROACH TO THE SLIDING MODE CONTROL OF PERIODIC REVIEW INVENTORY SYSTEMS 817

[19] W. Gao, Y. Wang, and A. Homaifa, “Discrete-time variable structure [36] C. Riddalls, S. Bennett, and N. Tipi, “Modelling the dynamics of supply
control systems,” IEEE Trans. Ind. Electron., vol. 42, no. 2, pp. chains,” Int. J. Syst. Sci., vol. 31, no. 8, pp. 969–976, 2000.
117–122, Apr. 1995. [37] H. Sarimveis, P. Patrinos, C. Tarantilis, and C. T. Kiranoudis, “Dy-
[20] G. Golo and Č. Milosavljević, “Robust discrete-time chattering free namic modeling and control of supply chain systems: A review,”
sliding mode control,” Syst. Control Lett., vol. 41, no. 1, pp. 19–28, Comput. Oper. Res., vol. 35, no. 11, pp. 3530–3561, 2008.
2000. [38] L. Song, S. Gong, and Y. Tian, “Study of reaching law approach for dis-
[21] K. Hoberg, J. Bradley, and U. Thonemann, “Analyzing the effect of the crete-time variable structure control system,” in Proc. IEEE Int. Conf.
inventory policy on order and inventory variability with linear control Modelling, Identif. Control, 2012, pp. 206–210.
theory,” Eur. J. Oper. Res., vol. 176, no. 3, pp. 1620–1642, 2007. [39] K. Subramanian, “Integration of control theory and scheduling
[22] P. Ignaciuk and A. Bartoszewicz, “LQ optimal and reaching law based methods for supply chain management,” Ph.D. dissertation, Univ.
sliding modes for inventory management systems,” Int. J. Syst. Sci., Wisconsin-Madison, Madison, WI, USA, 2013.
vol. 43, no. 1, pp. 105–116, 2012. [40] V. Utkin, “Variable structure systems with sliding modes,” IEEE Trans.
[23] P. Ignaciuk and A. Bartoszewicz, “Sliding mode dead-beat control of Autom. Control, vol. 22, no. 2, pp. 212–222, Apr. 1977.
perishable inventory systems with multiple suppliers,” IEEE Trans. [41] V. Utkin and S. V. Drakunow, “On discrete-time sliding mode control,”
Autom. Sci. Eng., vol. 9, no. 2, pp. 418–423, Apr. 2012. in Proc. IFAC Conf. Nonlinear Control, 1989, pp. 484–489.
[24] S. Janardhanan and B. Bandyopadhyay, “Output feedback sliding- [42] X. Yu and G. Chen, “Discretization behaviors of equivalent control
mode control for uncertain systems using fast output sampling tech- based variable structure systems,” IEEE Trans. Autom. Control, vol.
nique,” IEEE Trans. Ind. Electron., vol. 53, no. 4, pp. 1677–1682, 48, no. 9, pp. 1641–1646, Sep. 2003.
Oct. 2006. [43] X. Yu, B. Wang, Z. Galias, and G. Chen, “Discretization effect on
[25] S. Janardhanan and B. Bandyopadhyay, “Multirate output feedback equivalent control based multi-input sliding mode control systems,”
based robust quasi-sliding mode control of discrete-time systems,” IEEE Trans. Autom. Control, vol. 53, no. 6, pp. 1563–1569, Jul. 2008.
IEEE Trans. Autom. Control, vol. 52, no. 3, pp. 499–503, Mar. 2007. [44] X. Yu, B. Wang, and X. Li, “Computer-controlled variable structure
[26] S. Janardhanan and V. Kariwala, “Multirate output feedback based LQ systems: The state of the art,” IEEE Trans Ind. Informat., vol. 8, no. 2,
optimal discrete-time sliding mode control,” IEEE Trans. Autom. Con- pp. 197–205, May 2012.
trol, vol. 53, no. 1, pp. 367–373, Feb. 2008.
[27] I. Karaesmen, A. Scheller-Wolf, and B. Deniz, “Managing perishable Andrzej Bartoszewicz received the M.Sc. degree
and aging inventories: Review and future research directions,” in in 1987 and the Ph.D. degree in 1993, both from
Handbook of Production Planning, K. Kempf, P. Keskinocak, and R. the Łódź University of Technology, Łódź, Poland,
Uzsoy, Eds. Dordrecht, The Netherlands: Kluwer, 2008. and the Postdoctoral degree in control engineering
[28] S. Kurode, B. Bandyopadhyay, and P. Gandhi, “Discrete sliding mode and robotics from the Academy of Mining and
control for a class of underactuated systems,” in Proc. 37th Annual Metallurgy, Cracow, Poland.
Conf. IEEE Ind. Electron. Soc., 2011, pp. 3936–3941. He was a Visiting Scholar at Purdue University,
[29] A. Mehta and B. Bandyopadhyay, “Frequency-shaped sliding mode West Lafayette, IN, USA, and at Strathclyde Univer-
control using output sampled measurements,” IEEE Trans. Ind. Elec- sity, Glasgow, U.K. Then, for one year he was with
tron., vol. 56, no. 1, pp. 28–35, Jan. 2009. the University of Leicester, U.K. Currently, he is a
[30] A. Mehta and B. Bandyopadhyay, “The design and implementation of Professor with the Technical University of Łódź, a
output feedback based frequency shaped sliding mode controller for Vice Dean of the Faculty of Electrical, Electronic, Computer and Control Engi-
the smart structure,” in Proc. IEEE Int. Symp. Ind. Electron., 2010, pp. neering, the Head of the Electric Drive and Industrial Automation Group, and
353–358. a Vice-Director of the Institute of Automatic Control. He has published three
[31] S. Mija and T. Susy, “Reaching law based sliding mode control for monographs and over 250 papers, primarily in the field of sliding mode control.
discrete MIMO systems,” in Proc. IEEE Int. Conf. Control, Autom.,
Robot. Vision, 2010, pp. 1291–1296.
[32] Č. Milosavljević, “General conditions for the existence of a quasis-
liding mode on the switching hyperplane in discrete variable structure
systems,” Autom. Remote Control, vol. 46, no. 3, pp. 307–314, 1985. Piotr Leśniewski received the M.S. degree in control
[33] Č. Milosavljević, B. Peruničić-Draženović, B. Veselić, and D. Mitić, engineering and robotics from the Łódź University of
“Sampled data quasi-sliding mode control strategies,” in Proc. IEEE Technology, Łódź, Poland, in 2012. Currently, he is
Intern. Conf. Ind. Technol., 2006, pp. 2640–2645. working towards the Ph.D. degree at the Łódź Uni-
[34] Y. Niu, D. W. C. Ho, and Z. Wang, “Improved sliding mode control versity of Technology.
for discrete-time systems via reaching law,” IET Control Theory Appl., His main research interests include discrete-time
vol. 4, no. 11, pp. 2245–2251, 2010. sliding mode control and time-delay systems.
[35] Y. Pan and K. Furuta, “Variable structure control with sliding sector
based on hybrid switching law,” Int. J. Adaptive Control Signal
Process., vol. 21, pp. 764–778, 2007.