Applied Numerical Mathematics 37 (2001) 441458

Solving DDEs in M ATLAB

L.F. Shampine a, , S. Thompson b
a Mathematics Department, Southern Methodist University, Dallas, TX 75275, USA
b Department of Mathematics & Statistics, Radford University, Radford, VA 24142, USA

Abstract
We have written a program, dde23, to solve delay differential equations (DDEs) with constant delays in
M ATLAB. In this paper we discuss some of its features, including discontinuity tracking, iteration for short delays,
and event location. We also develop some theoretical results that underlie the solver, including convergence, error
estimation, and the effects of short delays on stability. Some examples illustrate the use of dde23 and show it to
be a capable DDE solver that is exceptionally easy to use. 2001 IMACS. Published by Elsevier Science B.V. All
rights reserved.

1. Introduction
Our goal is to make it as easy as possible to solve effectively a large class of delay differential equations
(DDEs). We restrict ourselves to systems of equations of the form

y  (x) = f x, y(x), y(x 1 ), y(x 2 ), . . . , y(x k )

(1)

for constant delays j such that = min(1 , . . . , k ) > 0. The equations are to hold on a
x
b, which
requires the history y(x) = S(x) to be given for x
a. Although DDEs with delays (lags) of more general
form are important, this is a large and useful class of DDEs. Indeed, Baker et al. [1] write that The lag
functions that arise most frequently in the modeling literature are constants. Restricting ourselves to this
class of DDEs makes possible a simpler user interface and more robust numerical solution.
A popular approach to solving DDEs exploits the fact that the task has much in common with the
solution of ordinary differential equations (ODEs). Indeed, the DDE solver dde23 that we discuss
here is closely related to the ODE solver ode23 from the M ATLAB ODE Suite [18]. In particular,
our attention is focused on the RungeKutta triple BS(2,3) used in ode23 because it has some special
properties that we exploit in dde23. (Refer to the survey by Zennaro [21] for a delightful discussion
of the important issues for more general methods of this type.) In Section 2 we explain how explicit
RungeKutta triples can be extended and used to solve DDEs. This extension is greatly simplified if the
* Corresponding author.

E-mail addresses: [email protected] (L.F. Shampine), [email protected] (S. Thompson).

0168-9274/01/$ see front matter 2001 IMACS. Published by Elsevier Science B.V. All rights reserved.
PII: S 0 1 6 8 - 9 2 7 4 ( 0 0 ) 0 0 0 5 5 - 6

442

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

maximum step size H is smaller than because the resulting formulas are explicit. This is no restriction
when proving convergence as H 0, so many papers assume it. However, we believe that it is not
acceptable in a quality solver, so we also discuss how to evaluate the implicit formulas that arise when
the step size is bigger than . In Section 3 we prove uniform convergence of the numerical scheme in
realistic circumstances. In the following section we justify an estimate of the local truncation error.
Generally y  (a+) from (1) is different from y  (a) = S  (a). It is characteristic of DDEs that this
discontinuity in y  propagates as a discontinuity in y  at a + 1 , . . . , a + k . In turn these discontinuities
propagate as a discontinuity in y (3) , and so forth. Locating and accounting for low-order discontinuities
in the solution is key to the use of RungeKutta methods for solving DDEs. In Section 5 we discuss the
practical issues of locating these discontinuities. Any DDE solver must account for them, but dde23
also deals with discontinuous changes in f at known points x in both the history and the interval of
integration in a similar way.
Some problems involve changes in f at points that depend on the solution itself, or more generally on
a function of the solution called an event function. Event functions are used to locate where y satisfies
certain conditions, e.g., where a specific component of y(x) has a local maximum. dde23 has a powerful
event location capability that is discussed in Section 6. It is difficult to deal with the many possibilities
that are seen in practice, so providing this capability affected profoundly the design of the solver.
The natural step size of a RungeKutta method is the largest that will yield the desired accuracy.
Delays that are long compared to this step size and delays that are short present special difficulties for a
DDE solver that we consider in Section 7. In this we prove a result about the stability of the numerical
scheme for small delays.
We close with a pair of examples that show restricting the class of problems, algorithmic developments,
language facilities in M ATLAB, and design have resulted in a DDE solver that is both capable and
exceptionally easy to use. We have prepared a tutorial that shows how to use dde23. The solver and
its auxiliary functions, the tutorial, and complete solutions of many examples from the literature are
available at http://www.runet.edu/~thompson/webddes/.

2. Formulas
Explicit RungeKutta triples are a standard way to solve the ODE problem y  = f (x, y) on [a, b]
with given y(a). They can be extended to solve DDEs. Indeed, dde23 is closely related to the ODE
solver ode23 [18] which implements the BS(2,3) triple [2]. A triple of s stages involves three formulas.
Suppose that we have an approximation yn to y(x) at xn and wish to compute an approximation at
xn+1 = xn + hn . For i = 1, . . . , s, the stages fni = f (xni , yni ) are defined in terms of xni = xn + ci hn and
yni = yn + hn

i1


aij fnj .

j =1

The approximation used to advance the integration is

yn+1 = yn + hn

s

i=1

bi fni .

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

443

For brevity we may write this in terms of the increment function

(xn , yn ) =

s


bi fni .

i=1

The solution satisfies this formula with a residual called the local truncation error, lten ,

y(xn+1 ) = y(xn ) + hn xn , y(xn ) + lten .

For sufficiently smooth f and y(x) this error is O(hp+1
n ). The triple includes another formula,

= yn + hn
yn+1

s


bi fni = yn + hn (xn , yn ).

i=1

The solution satisfies this formula with a local truncation error lten that is O(hpn ). This second formula is
used only for selecting the step size. The third formula has the form
yn+ = yn + hn

s


bi ( )fni .

i=1

The coefficients bi ( ) are polynomials in , so this represents a polynomial approximation to y(xn +

hn ) for 0

1. We assume that this formula yields the value yn when = 0 and yn+1 when = 1.
For this reason the third formula is described as a continuous extension of the first. A much more serious
assumption is that the order of the continuous extension is the same as that of the first formula. These
assumptions hold for the BS(2,3) triple. For such triples we regard the formula used to advance the
integration as just the special case = 1 of the continuous extension and write
yn+ = yn + hn (xn , yn , ).
The local truncation error of the continuous extension is defined by

y(xn + hn ) = y(xn ) + hn xn , y(xn ), + lten ( ).

for
We assume that for smooth f and y(x), there is a constant C1 such that lten ( )
C1 hp+1
n
0

1.
In his discussion of continuous extensions, Dormand [4] treats the local truncation error of the
BS(2,3) triple as an example. In Section 6.3 he obtains
1/2
2

1728 4 7536 3 + 13132 2 11148 + 3969
(2)
288
as a measure of the local truncation error at xn + hn . In Fig. 6.2 he shows that this measure increases
monotonely from zero at xn to a maximum at xn+1 . We shall see that the solver controls the size of
the error at xn+1 , so the continuous extension of this triple provides an accurate solution throughout
[xn , xn+1 ].
To use an explicit RungeKutta triple to solve the DDE (1), we need a strategy for handling the history
terms y(xni j ) that appear in

A(4) ( ) =

fni = f xni , yni , y(xni 1 ), . . . , y(xni k ) .

Two situations must be distinguished: hn
and hn > j for some j . We begin with the first. Suppose
that we have available an approximation S(x) to y(x) for all x
xn . If hn
, then all the xni j
xn
and

fni = f xni , yni , S(xni 1 ), . . . , S(xni k )

444

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

is an explicit recipe for the stage and the formulas are explicit. The function S(x) is the initial history
for x
a. After taking the step to xn+1 , we use the continuous extension to define S(x) on [xn , xn+1 ] as
S(xn + hn) = yn+ . We are then ready to take another step. This suffices for proving convergence as the
maximum step size tends to zero, but we must take up the other situation because in practice, we may very
well want to use a step size larger than the smallest delay. When hn > j for some j , the history term
S(x) is evaluated in the span of the current step and the formulas are defined implicitly. In this situation
we evaluate the formulas with simple iteration. On reaching xn , we have defined S(x) for x
xn . We
extend its definition somehow to (xn , xn + hn ] and call the resulting function S (0) (x). A typical stage
of simple iteration begins with the approximate solution S (m) (x). The next iterate is computed with the
explicit formula

S (m+1) (xn + hn) = yn + hn xn , yn , ; S (m) (x) .

Here and elsewhere it is convenient to show the dependence on the history by means of another argument
in the increment function. In the next section we show that if the step size is small enough, this is a
contraction and S(x) is well-defined for x
xn+1 .
In dde23 we predict S (0) (x) to be the constant y0 for the first step. We do not attempt to predict more
accurately then because the solution is not smooth at a and we do not yet know a step size appropriate
to the scale of the problem. Indeed, reducing the step size as needed to obtain convergence of simple
iteration is a valuable tactic for finding an initial step size that is on scale, cf. [16]. After the first step, we
use the continuous extension of the preceding step as S (0) (x) for the current step. This prediction has an
appropriate order of accuracy and we can even assess the accuracy quantitatively using (2). Remarkably,
the truncation error has a local maximum at xn+1 so that extrapolation actually provides a more accurate
solution for some distance. Specifically, the ratio of the error of yn+ to the error of yn+1 is no larger
than 1 for up to about 1.32. Much more important in practice is the accuracy of the prediction in the
span of a subsequent step of the same size or rather larger. Unfortunately, the ratio grows rapidly as
increases. Still, it is no bigger than 13 for 1

2, so the prediction is quite good when the step size
is changing slowly. The step size adjustment algorithms of dde23 are those of ode23 augmented to
(m+1)
(m)
yn+1
is no more
deal with implicit formulas. The convergence test for simple iteration is that yn+1
than one tenth the accuracy required of yn+1 . Because each iteration costs as much as an explicit step,
any proposed hn with < hn < 2 is reduced to so as to make the formulas explicit. Considering
the quality of the prediction when hn  2 , we have allowed up to 5 iterations. If convergence is not
achieved, hn is halved and the step repeated. The solver cannot crash because of repeated convergence
failures because eventually it will resort to a step size for which the formulas are explicit.

3. Convergence
Theorem. Suppose an explicit RungeKutta triple is used to solve the DDE (1) as described in the
preceding section. Assume that the meshes {xn } include all discontinuities of low orders and that the
maximum step size H satisfies (5) and (7). If f satisfies a Lipschitz condition in its dependent variables
and is sufficiently smooth in all its variables, then there exists a constant C such that for a
x
b,


y(x) S(x)
CH p .

(3)

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

445

This theorem shows how our method works. A more general convergence result may be found in [21].
Before proving this result, we begin our investigation of convergence by recalling the comments earlier
about how discontinuities arise and propagate for DDEs. We know in advance all the points where the
solution might not be smooth enough for the usual orders of the formulas to hold. Because we consider
one-step methods, if we take these points to be mesh points, the analysis is the same as if the solution
were smooth throughout the interval. Accordingly, we consider a sequence of meshes {xn } which include
all these points. Just how this is accomplished in practice is described in Section 5. Between these points
we assume that f and y(x) are smooth enough for the formulas to have their usual orders. We also
assume that f satisfies a Lipschitz condition with respect to all the dependent variables. For the sake of
simplicity, we suppose that there is only one delay and take the Lipschitz condition in the form





f (x, y,
z ) f (x, y, z)
L max y y , z z .

The RungeKutta formulas extended to DDEs involve a history term that we write generically in the
increment function as g(x). We require two lemmas about how the increment functions depend on their
arguments.
Lemma 1. There is a constant L such that for 0

1,







 xn , yn , ; g(x) xn , yn , ; g(x) 
L yn yn .

(4)

Proof. In a step of size hn from (xn , yn ), the intermediate quantities yni are defined by
i1


yni = yn + hn

aij f xnj , ynj , g(xnj )

j =1

and the yni are defined similarly in a step from (xn , yn ). Using the Lipschitz condition on f , it is
straightforward to show that for each i, there is a constant Li such that


yni yni
1 + H L

i1


Lj |aij | yn yn = Li yn yn .

j =1

With this result it follows easily that (4) holds with L = L

s

i=1 Li

max(|bi ( )|).

The second lemma we need is a bound on the effect of using different history terms.
Lemma 2. Let be a bound on G(x) g(x) for all x
xn+1 . If the maximum step size H is small
enough that
 i1


H L max
i

|aij |
1

(5)

j =1

then there is a constant such that







 xn , yn , ; G(x) xn , yn , ; g(x) 
.

Proof. Let the intermediate quantities of the two formulas be

rni = yn + hn

i1

j =1

aij f xnj , rnj , G(xnj )

(6)

446

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

and
sni = yn + hn

i1


aij f xnj , snj , g(xnj ) .

j =1

The Lipschitz condition on f , inequality (5), and a simple induction argument show that for each i,
rni sni
hn

i1


|aij |L
.

j =1

With this result it follows easily that (6) holds with = L si=1 max(|bi ( )|).
Recall that if the step size is small enough, the intermediate results yni are defined by the explicit
formulas
yni = yn + hn

i1


aij f xnj , ynj , S(xnj ) ,

j =1

but if hn > , we may have some xnj j > xn so that S(x) is evaluated in the span of the current step. In
this situation we evaluate the formulas with simple iteration as described in the preceding section. Using
this lemma we obtain a condition on the maximum step size that insures the convergence of simple
iteration: Let (m) be the maximum of S (m+1) (x) S (m) (x) for xn
x
xn+1 . (The functions are the
same for x
xn .) Inequality (6) implies that (m)
hn (m1)
H (m1) . From this it is obvious
that simple iteration is contracting and the formula is well-defined if the maximum step size H is small
enough to satisfy both (5) and
H = HL

s


 1
max |bi ( )|
.
2
i=1

(7)

We are now in a position to prove convergence as H 0 when the meshes include all discontinuities
of low order and H satisfies (5) and (7). Suppose that the integration has reached xn and let En be a
bound on y(x) S(x) for all x
xn . The local truncation error of the continuous extension is defined
by

y(xn + hn ) = y(xn ) + hn xn , y(xn ), ; y(x) + lten ( ).

If we introduce

wn+ = yn + hn xn , yn , ; y(x)

then inequality (4) and our assumption about the local truncation error imply that for 0

1,



 

y(xn + hn ) wn+ 
(1 + hn L)y(xn ) yn  + lten ( )

(1 + hn L)En + C1 hp+1
n .
The numerical solution is

yn+ = yn + hn xn , yn , ; S(x) .
The next step of the proof is complicated by the possibility that this formula is implicit. If we let be
the maximum of y(x) S(x) for x
xn+1 , then inequality (6) states that


yn+ wn+ = hn  xn , yn , ; S(x) xn , yn , ; y(x)
hn

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

447

and the triangle inequality implies that



y(xn + hn ) yn+ 
(1 + hn L)En + C1 hp+1 + hn .
n

This is a bound on y(x) S(x) for xn
x
xn+1 and by definition of En , the bound also holds for all
x
xn . We conclude that
+ hn .

(1 + hn L)En + C1 hp+1
n
With the assumption that H satisfies (7), it then follows that



C1
1 + hn L
En +
hp+1

1 + 2(L + )hn En + 2C1 hp+1


n
n .
1 hn
1 hn
If we let C2 = 2(L + ) and C3 = 2C1 , this inequality tells us that
En+1 = (1 + hn C2 )En + C3 hp+1
n
is a bound on y(x) S(x) for all x
xn+1 . Because we start with the given initial history, E0 = 0.
A simple induction argument shows that for all n,
En
C3 eC2 (xn a) (xn a)H p .
The uniform bound
En
C3 eC2 (ba) (b a)H p = CH p
implies (3) for a
x
b.
4. Error estimation and control
Nowadays codes based on explicit RungeKutta triples estimate and control the error made in the
current step by the lower order formula. They advance the integration with the higher order result yn+1
(local extrapolation) because it is more accurate, though just how much is not known. Our notation and
proof of convergence incorporate this assumption about the triple. We make use of local extrapolation
in our proof that we can estimate the local truncation error of the lower order formula, lten , by the

.
computable quantity est = yn+1 yn+1
The local truncation error

lten = y(xn+1 ) y(xn ) hn xn , y(xn ); y(x)

is O(hpn ). Using the corresponding definition of lten it follows easily that

.
lten = hn xn , y(xn ); y(x) hn xn , y(xn ); y(x) + O hp+1
n
From the definition of yn+1 we have

hn xn , yn ; S(x) = yn+1 yn .
Using the convergence result (3), we show now that

hn xn , y(xn ); y(x) = hn xn , yn ; S(x) + O hn H p .

First we use inequality (4) to see that










hn xn , y(xn ); y(x) hn xn , yn ; y(x) 
hn Ly(xn ) yn 
hn LCH p

448

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

and then inequality (6) to see that








hn xn , yn ; y(x) hn xn , yn ; S(x) 
hn CH p .

The desired relation follows from the triangle inequality. The same argument establishes the corresponding relation for the lower order formula. With these two relations we have

lten = hn xn , yn ; S(x) hn xn , yn ; S(x) + O hn H p

= (yn+1 yn ) yn+1
yn + O hn H p

= est + O hn H p

which justifies our estimate of the local truncation error.

As with ODEs, the local truncation error can be expanded to

.
lten = xn , y(xn ); y(x) hpn + O hp+1
n
If we reject the step of size hn from (xn , yn ), we predict the local truncation error of another attempt of
size h to be

est(h/ hn)p xn , y(xn ); y(x) hp .

The same prediction applies to a step of size h from (xn+1 , yn+1 ) on making the approximation
(xn+1 , y(xn+1 ); y(x)) (xn , y(xn ); y(x)) because the change in each of the arguments is O(hn ).

, but advance the integration with yn+1 because it is

We control the local truncation error of yn+1
believed to be more accurate. Recall that for the BS(2,3) triple, the local truncation error of yn+ attains
its maximum for yn+1 . Accordingly, though we do not know the local truncation error of the continuous
extension, we believe that it is smaller than required throughout the span of the step.

5. Tracking discontinuities
The constant lags 1 , . . . , k are supplied to dde23 as the array lags. It is required that =
min(1 , . . . , k ) > 0 and that the delays are distinct. In simplest use the only discontinuity is at the
initial point where the derivative obtained from the DDE, y  (a+) = f (a, y(a), y(a 1 ), y(a 2 ), . . . ,
y(a k )), is generally different from that provided by the history, S  (a). Generally the solution itself
is continuous at the initial point, so we defer consideration of the case when it is not. It is characteristic
of DDEs that this discontinuity in the first derivative results in discontinuities in the second derivative at
a + 1 , . . . , a + k . We describe this as discontinuities at the first level. In turn each of these discontinuities
propagates in the next level as a discontinuity in the third derivative, and so forth. This is much simpler
than the situation with more general DDEs because we can determine easily, and in advance, where the
solution might have discontinuities and the lowest order discontinuity possible there. Some solvers ask
users to supply details about which y(x j ) appear in which equations so as to determine more precisely
how discontinuities propagate. We think that the inconvenience of this design outweighs any advantages
gained in the solution of the DDEs. Moreover, assuming that the worst can happen is more robust because
there is no opportunity for a user to make a mistake in specifying the structure of the DDEs.
We have to track discontinuities and account for them during the integration because the usual
order of a RungeKutta formula is seen in a step of size hn from xn only when f (x, y(x), y(x 1 ),
y(x 2 ), . . . , y(x k )) is sufficiently smooth. As we have described the solution of DDEs with explicit

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

449

RungeKutta methods, each step is taken with two formulas in order to estimate the error made in the
step. If the error estimation and step size adjustment algorithms are to work properly, f must be smooth
enough for both to have their expected order. If we require that the step size be chosen so that each point
of discontinuity is a mesh point xm , then none of y(x), y(x 1 ), . . . , y(x k ) will have a low order
discontinuity in the span of the current step. This is obvious for a discontinuity in the argument y(x).
There cannot be a in (xn , xn + hn ) for which some y( j ) is not smooth because the discontinuity in
y at j would have propagated to and we would have limited hn so that xn + hn
. Because we
consider only one-step methods, it is only the smoothness of f in the current step that affects the order
of the formula. This is true even if y or f is discontinuous at xn provided that the correct values are used
at xn +, but we need take no special action in the solver because with one exception that we take up later,
we assume these functions are (at least) continuous.
Although constant delays simplify greatly the location of potential discontinuities, there are some
subtleties. One is revealed by a simple example. Suppose we start the integration at x = 0 and the delays
are 1/3 and 1. The first lag implies a potentially severe lack of smoothness at 1/3, 2/3, 3/3, . . . . The
difficulty is that in finite precision arithmetic, representation of 1/3 is not exact, implying that the
propagated value that ought to be 1 is merely very close to the correct value. Generally this is not
important, but the lag of 1 implies a discontinuity precisely at 1. We must recognize that finite precision
arithmetic has split two values that are the same else the requirement that the solver step to points of
discontinuity will cause it to attempt steps that are too short to be meaningful in the precision available.
In dde23 we purge one of any pair of values that differ by no more than ten units of roundoff. This is
done at each level of propagation so as to remove duplicates as soon as possible.
Example 4.4 of Oberle and Pesch [13] is a model for the spread of an infection due to Hoppensteadt
and Waltman. The problem is posed on [0, 10] and there is one lag, = 1. The solution is continuous
throughout the interval, but there are discontinuous changes in the differential equation at points known
in advance. Indeed, the changes are qualitative because at first the equation does not depend on y(x 1),
hence is an ODE. We provide in dde23 for discontinuities at points known in advance. All the user need
do is give the locations of potential discontinuities as the value of the option Jumps. As with the ODE
Suite, options for dde23 are set with an auxiliary function, ddeset. In the case of the HoppensteadtWaltman model, the three points where discontinuities occur are defined by
c = 1/sqrt(2);
options = ddeset(Jumps,[(1-c), 1, (2-c)]);
and conveyed to the solver by means of the optional argument options. Discontinuities at known
points are not limited to the interval of integration. The Marchuk immunology model discussed in [7,
pp. 297298] is solved on [0, 60] and there is one lag, = 0.5. One component of the history function
is V (t) = max(0, t + 106 ), which has a discontinuity in its first derivative at t = 106 . With the user
interface of dde23 it is easy to specify discontinuities at known points and they are accommodated in
the algorithm simply by forming an array that includes these points as well as the initial point and then
propagating them all as described already for the initial point. When using a solver with the capabilities of
dde23 it is easy and certainly better practice to specify discontinuities like those of these two examples,
but a robust integrator may well be able to cope with them automatically.
Much more serious complications are introduced by continuation (restarts). As we discuss more fully
in Section 6, we provide for the location of events. This capability is used, for example, to deal with
a coefficient (m) in the Marchuk model that has a jump in its first derivative with respect to the state

450

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

variable m(t). Restarting the integration when this happens assures us that the formulas are applied only
to smooth functions. Retaining the solution in the form of a structure makes it possible to continue. For
the present we note only that we must include in the solution structure the list of potential discontinuities.
After all, we might have discarded points outside the original interval of integration that lie in the span of
the current interval. Accordingly, on a restart this list, any jumps specified by the user in the current
integration, and the initial point of the current integration are formed into an array. The entries are
propagated and duplicates purged as described earlier.
An initial discontinuity in the first derivative appears in derivatives of successively higher order. When
the discontinuities are in a derivative of sufficiently high order that they do not affect the order of the
formulas, we can stop tracking them. The relatively low order of the BS(2,3) triple implies that the effects
of discontinuities are rather more short-lived than with other popular formulas. Our basic assumption is
that y(x) is continuous, so only four levels of propagation are needed. During the integration itself, the
only time we permit a discontinuity in y(x) is at the initial point, which is indicated by providing the
value of the solution as the value of the option InitialY. The DDE is evaluated using this value,
so both y and f are continuous to the right of the initial point. The facilities of dde23 allow us to
deal with discontinuous changes in the solution at other times. If the discontinuity occurs during the
integration, as for example at the time of occurrence of an event, the integration is to be restarted with the
appropriate value of the solution supplied via InitialY. Potential discontinuities of any order in the
initial history are specified via the Jumps option. If either kind of discontinuity in y(x) is possible,
discontinuities are propagated to one higher level in the solver.
The HoppensteadtWaltman model involves the solution of ODEs as well as DDEs. The solver makes
no assumption about whether terms with lags actually appear in the equations. This makes it possible
to solve ODEs, though it is best then to set lags = [] because any delay specified affects the list of
potential low-order discontinuities, hence the details of the integration.

6. Event location
Recall that the Marchuk model has a coefficient (m) that has a jump in its first derivative with respect
to the state variable m(t). Such discontinuities are qualitatively different from the discontinuities that we
treat with the Jumps option because they occur at unknown times. An event is said to occur at time
t when a scalar function g(t, y(t), y(t 1 ), . . . , y(t k )), called an event function, vanishes at t = t .
There may be many event functions and it may be important how the function vanishes, but we defer
discussion of these complications. As with the propagated discontinuities, if we locate events and restart
there, we integrate only with smooth functions. The theoretical and practical difficulties of event location
are often not appreciated, but a theoretical analysis is possible with reasonable assumptions, e.g. [19],
and a number of quality ODE solvers provide for the location of events. In particular, all the solvers of
the M ATLAB ODE Suite have a powerful event location capability that we have exploited in developing
dde23. The capability is realized by means of the zero-finding function odezero. A nice exposition of
its algorithm is found in [11]. Although we did not change the basic algorithm, we modified the function
to account for the circumstances of dde23.
The user interface for event location in dde23 is similar to that of ode23. There is no limit on the
number of scalar event functions. They are all evaluated in a single function and the values returned as a
vector. Using ddeset, the name of this function is passed to the solver as the value of the Events

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

451

option. The function must also return some information about the nature of the events. Sometimes a user
wants to know when events occur and the solution then, e.g., where a solution component has its local
maxima and its values there. This is quite different from the events of the Marchuk model which cause
the integration to be terminated. The two situations are distinguished using a vector isterminal. If we
want to terminate the integration when event function k vanishes, we set component k of isterminal
to 1, and otherwise to 0. There is an annoying matter of some importance: Sometimes we want to start
an integration with an event function that vanishes at the initial point. Imagine, for example, that we fire
a model rocket into the air and we want to know when it hits the ground. It is natural to use the height of
the rocket as a terminal event function, but it vanishes at the initial time as well as the final time. dde23
treats an event at the initial point in a special way. The solver locates such an event and reports it, but does
not treat it as terminal, no matter how isterminal is set. It may be important how an event function
vanishes. For example, to find local maxima of a solution component, we can locate zeros of the first
derivative of this component. However, to distinguish maxima from minima, we want the solver to report
a zero only when this function decreases through 0. This is done using the vector direction. If we
are interested only in events for which event function k is increasing through 0, we set component k of
direction to +1. Correspondingly, we set it to 1 if we are interested only in those events for which
the event function is decreasing, and 0 if we are interested in all events.
It is hard to devise a good interface for event location because users may want to do quite different
things when an event occurs. We have distinguished between terminal and non-terminal events. dde23
returns its solution in the form of a structure. It can be called anything, but to be specific, suppose it
is called sol. Often a user will want to know when an event occurred, what the solution is there, and
which event function vanished. This information is returned by means of fields that have fixed names. If
there are events, they occur at the entries of the vector sol.xe. For each entry, the solution there is the
corresponding column of the array sol.ye. Also, the corresponding entry of the vector sol.ie is the
index of the event function that vanished.
Often an event is terminal because it is necessary to change the function defining the equations, as with
the Marchuk model, and/or the value of the solution before continuing the integration. The user interface
for event location in ode23 is powerful and reasonably convenient when solving ODEs because on a
return from a terminal event, it is not difficult to make such changes, continue on as the solution of a new
problem, and finally assemble the solutions on subintervals to obtain a solution on the whole interval of
interest. The matter is much more difficult for DDEs because they involve previously given or computed
quantities. On a restart, the solver may have to evaluate the given history function, which can be specified
either as a function or a vector, as well as evaluate the solution computed prior to the restart. As mentioned
in Section 5, propagated discontinuities also present complications. We deal with restarts by allowing the
history to be specified as a solution structure. This structure contains all the information the solver needs
to recognize and deal with the various possibilities. The approach is easy for the user, indeed, easier than
when solving ODEs because the solution structure always contains the solution from the initial point to
the last point reached in the integration.
As stated earlier, discontinuities in the solution are allowed only at the initial point of an integration.
The solver is told of this and given the value of the solution there by providing it as the value of the option
InitialY. Discontinuities in the solution are not common, and when they do occur, they invariably
do so when a suitably defined event function vanishes. By making the event terminal, we can use the
InitialY option to give the solution the correct initial value for continuation.

452

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

7. Long and short delays

The natural step size is the largest one that yields the specified accuracy. Codes for the solution of
DDEs have difficulties when the delays are both long and short compared to the natural step size. Long
delays present no special difficulty for dde23, so after we explain why, we give our attention to short
delays.
On reaching a point xn , the solver needs to evaluate the solution at previous points xn j , hence must
retain the information it needs to do this. When there is a delay long compared to the natural step size, this
implies that the solution must be retained over many steps. Many solvers are written in FORTRAN 77.
Because this environment requires allocation of storage at the beginning of a run, there is a possibility of
exceeding the storage provided. This is not an issue for dde23 because we have followed the traditional
design of ODE solvers in M ATLAB of returning to the user the solution at all mesh points. Virtual storage
and dynamic memory management increase the class of problems that can be solved, but we assume that
long delays will not prevent solution of the problem given.
An ODE problem is stiff when the natural step size must be reduced greatly and often. Although
the reduction necessary for the stability of explicit formulas gets the most attention, [16] points out
reductions arising in the evaluation of implicit formulas by simple iteration and in output that might be
just as severe. dde23 uses simple iteration to evaluate the implicit formulas that arise when is smaller
than the step size. The sufficient condition (7) for the convergence of simple iteration does not depend
on , so delays cannot restrict the step size greatly for this reason. Some DDE solvers require that H <
so as to have the simplicity of evaluating explicit formulas. For such solvers a short delay does cause
stiffness. Another issue specific to DDEs is that solvers like dde23 do not step over discontinuities. A
short delay may cause the natural step size to be reduced greatly for this reason. However, this is not
likely to happen so often when using dde23 as to constitute stiffness. For one thing, the relatively low
order of the formulas limits the number of steps affected by propagated discontinuities. For another, not
stepping over discontinuities is different from limiting the step size to the shortest delay. To appreciate the
distinction, suppose that there are two delays, 1  2 . The first few steps of the integration are limited to
1 because there are potential discontinuities of low order at a + 1 , a + 21 , . . .. The other delay makes
its appearance first at a + 2 . Once the discontinuity at a that is propagating at spacing 1 has smoothed
out, the solver can use step sizes much bigger than 1 in the rest of the integration to a + 2 . This all
repeats at a + 2 because the discontinuity there propagates at a spacing of 1 for a few steps and the
second delay has its next effect at a + 22 .
There is an extensive literature on the stability of RungeKutta methods for DDEs, cf. [6,8,14,15] and
references cited therein, much of it concerned with solving the scalar model problem
y  (x) = Ly(x) + My(x )

(8)

with constant step size h. Our investigation is somewhat novel in that we consider short delays and we
relate the stability of our scheme when solving the model equation for DDEs to its stability when solving
the test equation for ODEs, y  (x) = Ly(x). Specifically, we prove that if the scheme is stable in a strong
sense for the test equation, then for all sufficiently small and M, it is stable for (8). We conclude that
in interesting circumstances, small delays do not affect adversely the stability of the scheme.
Suppose the integration has reached xn and we take a step of size h. For all sufficiently small , the
only argument of the BS(2,3) triple that does not lie in the span of the current step is xn . It lies

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

453

in [xn h, xn ] where S(x) is the cubic Hermite interpolant to yn1 , fn1 , yn , fn . As the form of the
interpolant suggests and Maple [9] confirms,

S(xn ) = yn fn + O 2 .

(9)

The only way that yn1 and fn1 affect the computation of yn+1 is through the value S(xn ). Because
these quantities from the preceding step do not appear in the lowest order terms of the expansion of
S(xn ), we can investigate stability in the limit 0 by considering only the current step. With this
in mind, we can normalize by taking yn = 1 so that for small , the integration is stable if |yn+1 | < 1.
In what follows we work only with terms through O( 2 ). The quantity fn is defined implicitly because
fn = Lyn + MS(xn ) and S(xn ) depends on fn . A little calculation using (9) shows that



L+M
fn =
yn + O 2 .
M + 1
A similar expression holds for fn+1 . Using Maple it is now straightforward to determine how yn+1
depends on h, L, M, . Examination of the expression shows that it is easier to interpret if we introduce
z = hL and Z = hM.
Insight and a check on the computations is provided by supposing that M = 0. In this situation yn+1
should be, and is, a third order approximation to the solution of the test equation with yn = 1, namely
yn+1 = P (z) = 1 + z + z2 /2 + z3 /6. The polynomial P (z) appearing here is the stability polynomial
of the method for ODEs. The method is stable if |P (z)|
1 and we shall say that it is damped if
|P (z)| < 1.
So far we have been considering an expression for yn+1 that is useful for small . Let us now suppose
that Z is also small. A straightforward computation using Maple shows that




w20 (z)
+ O Z 2 + O( Z) + O 2 .
h
This expression makes clear that if for given h and L, the method is damped for the test equation, then it
is stable for the model DDE (8) for all sufficiently small and M.

yn+1 = P (z) + w12 (z) Z +

8. Examples
In this section we consider a pair of examples that show how easily even rather complicated DDEs can
be solved with dde23.
8.1. Example 1
A KermackMcKendrick model of an infectious disease with periodic outbreak is discussed in [7]
where Fig. 15.6 shows the solution of
y1 (t) = y1 (t)y2 (t 1) + y2 (t 10),
y2 (t) = y1 (t)y2 (t 1) y2 (t),
y3 (t) = y2 (t) y2 (t 10)
on [0, 40] with history y1 (t) = 5, y2 (t) = 0.1, y3 (t) = 1 for t
0.

(10)

454

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

Solving this model illustrates the simplest use of dde23. In general a call to the solver with all options
set by default has the form
sol = dde23(ddefile,lags,history,tspan);
The call list could scarcely be shorter. The interval of integration, tspan, is here [0, 40]. history
can be specified in several ways. In the common case that the solution is constant prior to the initial point,
the vector itself can be supplied as history. For the example it is [5; 0.1; 1]. The initial value of
the solution is not an input argument for dde23 because most problems have initial values that are the
same as those given by the history function evaluated at the initial point. The constant delays are supplied
as an array lags, here [1, 10]. There is no limit on the number of delays, but they must be distinct.
ddefile is the name of the function for evaluating the DDEs. For the example this function can be
coded as
function v = kmf(t,y,Z)
ylag1 = Z(:,1);
ylag2 = Z(:,2);
v = zeros(3,1);
v(1) = - y(1)*ylag1(2) + ylag2(2);
v(2) = y(1)*ylag1(2) - y(2);
v(3) = y(2) - ylag2(2);
Here the vector y approximates y(t) and column j of the array Z approximates y(t j ) for j = 1, . . . , k.
The delays can be specified in any order, but in defining the DDEs, j is lags(j). It is not necessary
to create a local vector ylag1 approximating y(t 1 ) as we have done here, but we find that it often
makes the equations easier to read.
Having coded the DDEs, we now solve the initial value problem with the command
sol = dde23(kmf,[1, 10],[5; 0.1; 1],[0, 40]);
The input arguments of dde23 are much like those of ode23, but the output differs formally in that it
is one structure, here called sol, rather than several arrays
[t,y,...] = ode23(...
The field sol.x corresponds to the array t of values of the independent variable returned by ode23
and the field sol.y, to the array y of solution values. Fig. 15.6 of [7] is reproduced by
plot(sol.x,sol.y);
In summary, a complete program for solving the KermackMcKendrick model consists of the file kmf.m
defining the DDEs, the one command invoking dde23, and the one plot command.
Output from dde23 is not just formally different from that of ode23. The method of dde23
approximates y(t) by a piecewise-polynomial function S(t) C 1 [a, b]. The solver places in sol the
information necessary to evaluate this function with ddeval. Values of S(t) and optionally S  (t) are
obtained at an array of arguments t by
[S,Sp] = ddeval(sol,t);

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

455

With this form of output, we solve a DDE just once and then obtain inexpensively as many solution values
as we like, anywhere we like. Plotting the solution at the mesh points of sol.x is generally satisfactory,
but when the graph is not sufficiently smooth, we can always get a smooth graph by evaluating it at
enough points with ddeval.
To further illustrate the advantages of output in this form, we explain how to reproduce a figure in [5].
A DDE with a single lag of 14 is solved there on [0, 300]. Fig. 2a plots y(t 14) against y(t), but starts
with t = 50 so as to allow transient behavior to die out. The problem is solved easily with dde23, but we
cannot make this plot using results at the mesh points sol.x alone. This is because they are not equally
spaced: If t appears in sol.x so that we have y(t ) available in sol.y, it is generally not the case that
t 14 appears in sol.x, so we do not have y(t 14). However, using ddeval we can reproduce the
figure easily with
t = linspace(50,300,1000);
y = ddeval(sol,t);
ylag = ddeval(sol,t - 14);
plot(y,ylag);
We specify 1000 points because generally a smooth graph in the phase plane requires a good many.
Because ddeval is vectorized and exploits fast built-in functions, there is no need to be concerned
about evaluating the solution at so many points.
As noted in Section 5, dde23 does not require that the equations actually involve the delays specified.
By specifying an additional, artificial delay that is short, the way the solution is computed is affected, but
the solution and the natural step size remain the same. Solving the KermackMcKendrick model with the
option Stats set on, it is found that the problem is solved in 133 successful steps, with 17 failed
attempts and a total of 451 function evaluations. Adding an artificial delay of 104 by changing lags
to [1, 10, 1e-4] increased the number of successful steps to 164, the number of failed attempts
dropped to 12, and the total number of function evaluations rose to 1027. If the step size had been
restricted to the shortest delay, the integration would have required at least 40 104 steps!
8.2. Example 2
A two-wheeled suitcase may begin to rock from side to side as it is pulled. When this happens, the
person pulling it attempts to return it to the vertical by applying a restoring moment to the handle. There
is a delay in this response that can affect significantly the stability of the motion. This is modeled by
Suherman et al. [20] with the DDE

 (t) + sign (t) cos (t) sin (t) + (t ) = A sin(1t + ).

(11)

The equation is solved on [0, 12] as a pair of first order equations with y1 = , y2 =  . Fig. 3 of [20]
shows a plot of y1 (t) against t and a plot of y2 (t) against y1 (t) when = 2.48, = 1, = 0.1, A =
0.75, 1 = 1.37, = arcsin( /A) and the initial history is the constant vector zero.
This complicated problem can be solved with a short program that we state now and then discuss:
state = +1;
opts = ddeset(RelTol,1e-5,AbsTol,1e-5,Events,scasee);
sol = dde23(scasef,0.1,[0; 0],[0, 12],opts,state);

456

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

while sol.x(end) < 12

if sol.ie(end) == 1
% A wheel hit the ground.
state = - state;
opts = ddeset(opts,InitialY,[0; 0.913*sol.y(2,end)]);
sol = dde23(scasef,0.1,sol,[sol.x(end), 12],opts,state);
else
% The suitcase fell over.
break;
end
end
A parameter state is used for the value of sign(y1 (t)). Parameters can be communicated as global
variables or by adding them at the ends of the call lists. Using the latter technique, the ddefile can be
coded as
function v = scasef(t,y,Z,state)
ylag = Z(:,1);
v = [ y(2); 0];
eta = asin(2.48/0.75);
v(2) = sin(y(1)) - state*0.248*cos(y(1)) - ylag(1) ...
+ 0.75*sin(1.37*t + eta);
A wheel hits the ground (the suitcase is vertical) when y1 (t) = 0 and the suitcase has fallen over when
|y1 (t)| = /2. All these events are terminal and all are to be reported. The event function can be coded
as
function [value,isterminal,direction] = scasee(t,y,Z,state)
value = [y(1); abs(y(1))-pi/2];
isterminal = [1; 1];
direction = [0; 0];
When a wheel hits the ground, the integration is to be restarted with y1 (t) = 0 and y2 (t) multiplied by
the coefficient of restitution 0.913. The InitialY option is used for this purpose. The solution at
all the mesh points is available as the field sol.y and in particular, the solution at the time of a terminal
event is the last column of this array, sol.y(:,end). The sign of state is changed to reflect the
change in sign of y1 (t). If the suitcase falls over, the run is terminated, so we must check which event
occurred. This information is provided in sol.ie and it is the last event that we must check. Because
we do not know how many events will occur, restarts are coded in a while loop.
Options are set by means of the auxiliary function ddeset exactly as with odeset when using
ode23. In the first call to ddeset, we specify the event function and error tolerances. The call inside
the while loop shows how to add an option and change the value of an option previously set.
The program reproduces Fig. 3 of [20] except that we have not included the standard commands
that plot the figures. The event locations computed by dde23 are compared in Table 1 to reference
values computed with the FORTRAN 77 code DKLAG5 [12] used in [20] and verified with its successor
DKLAG6 [3]. Having written the three solvers, we can fairly say that it is very much easier to solve this
problem in M ATLAB with dde23.

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

457

Table 1
Events for the suitcase problem
dde23

reference

A wheel hit the ground at

4.5168

4.516757

A wheel hit the ground at

9.7511

9.751053

11.6704

11.670393

The suitcase fell over at

After running this program, sol.xe is

0.0000

4.5168

4.5168

9.7511

9.7511

11.6704

This does not seem to agree with the event locations of Table 1. For instance, why is there an event at 0?
That is because one of the event functions is y1 and this component of the solution has initial value 0.
As explained in Section 6, dde23 locates this event, but does not terminate the integration because the
terminal event occurs at the initial point. The first integration terminates at the first point after the initial
point where y1 (t ) = 0, namely t = 4.5168. The second appearance of 4.5168 in sol.xe is the same
event at the initial point of the second integration. The same thing happens at 9.7511 and finally the event
at 11.6704 tells us that the suitcase fell over and we are finished.

References
[1] C.T.H. Baker, C.A.H. Paul, D.R. Will, Issues in the numerical solution of evolutionary delay differential
equations, Adv. Comput. Math. 3 (1995) 171196.
[2] P. Bogacki, L.F. Shampine, A 3(2) pair of RungeKutta formulas, Appl. Math. Lett. 2 (1989) 321325.
[3] S.P. Corwin, D. Sarafyan, S. Thompson, DKLAG6: A code based on continuously imbedded sixth order
RungeKutta methods for the solution of state dependent functional differential equations, Appl. Numer.
Math. 24 (1997) 319333.
[4] J.R. Dormand, Numerical Methods for Differential Equations, CRC Press, Boca Raton, FL, 1996.
[5] J.D. Farmer, Chaotic attractors of an infinite-dimensional dynamical system, Phys. D 4 (1982) 366393.
[6] N. Guglielmi, E. Hairer, Order stars and stability for delay differential equations, Numer. Math. 83 (1999)
371383.
[7] E. Hairer, S.P. Nrsett, G. Wanner, Solving Ordinary Differential Equations I, Springer-Verlag, Berlin, 1987.
[8] K.J. int Hout, On the stability of adaptations of RungeKutta methods to systems of delay differential
equations, Appl. Numer. Math. 22 (1996) 237250.
[9] Maple V Release 5, Waterloo Maple Inc., 450 Phillip St., Waterloo, ON N2L 5J2, Canada, 1998.
[10] M ATLAB 5, The MathWorks, Inc., 3 Apple Hill Dr., Natick, MA 01760, 1998.
[11] C.B. Moler, Are we there yet?Zero crossing and event handling for differential equations, M ATLAB News
& Notes (1997) 1617, available at http://www.mathworks.com/company/newsletter/pdf/97slCleve.pdf.
[12] K.W. Neves, S. Thompson, Software for the numerical solution of systems of functional differential equations
with state dependent delays, Appl. Numer. Math. 9 (1992) 385401.
[13] H.J. Oberle, H.J. Pesch, Numerical treatment of delay differential equations by Hermite interpolation, Numer.
Math. 37 (1981) 235255.
[14] C.A.H. Paul, C.T.H. Baker, Stability boundaries revisited-RungeKutta methods for delay differential
equations, Numer. Anal. Rept. No. 205 (revised), Math. Dept., University of Manchester, UK, 1991.

458

L.F. Shampine, S. Thompson / Applied Numerical Mathematics 37 (2001) 441458

[15] C.A.H. Paul, C.T.H. Baker, Explicit RungeKutta methods for the numerical solution of singular delay
differential equations, Numer. Anal. Rept. No. 212 (revised), Math. Dept., University of Manchester, UK,
1992.
[16] L.F. Shampine, Numerical Solution of Ordinary Differential Equations, Chapman & Hall, New York, 1994.
[17] L.F. Shampine, Interpolation for RungeKutta methods, SIAM J. Numer. Anal. 22 (1985) 10141027.
[18] L.F. Shampine, M.W. Reichelt, The M ATLAB ODE suite, SIAM J. Sci. Comput. 18 (1997) 122.
[19] L.F. Shampine, S. Thompson, Event location for ordinary differential equations, Comput. Math. Appl. 39
(2000) 4354.
[20] S. Suherman, R.H. Plaut, L.T. Watson, S. Thompson, Effect of human response time on rocking instability of
a two-wheeled suitcase, J. Sound Vibration 207 (1997).
[21] M. Zennaro, Delay differential equations: Theory and numerics, in: M. Ainsworth et al. (Eds.), Advances in
Numerical Analysis: Theory and Numerics of Ordinary and Partial Differential Equations, Vol. IV, Clarendon
Press, Oxford, 1995, pp. 291333.