NAG Library Routine Document
E05UCF
1 Purpose
E05UCF is designed to find the global minimum of an arbitrary smooth function subject to constraints (which may include simple bounds on the variables, linear constraints and smooth nonlinear constraints) by generating a number of different starting points and performing a local search from each using sequential quadratic programming.
2 Specification
SUBROUTINE E05UCF ( 
N, NCLIN, NCNLN, A, LDA, BL, BU, CONFUN, OBJFUN, NPTS, X, LDX, START, REPEAT, NB, OBJF, OBJGRD, LDOBJD, ITER, C, LDC, CJAC, LDCJAC, SDCJAC, R, LDR, SDR, CLAMDA, LDCLDA, ISTATE, LISTAT, IOPTS, OPTS, IUSER, RUSER, INFO, IFAIL) 
INTEGER 
N, NCLIN, NCNLN, LDA, NPTS, LDX, NB, LDOBJD, ITER(NB), LDC, LDCJAC, SDCJAC, LDR, SDR, LDCLDA, ISTATE(LISTAT,NB), LISTAT, IOPTS(740), IUSER(*), INFO(NB), IFAIL 
REAL (KIND=nag_wp) 
A(LDA,*), BL(N+NCLIN+NCNLN), BU(N+NCLIN+NCNLN), X(LDX,NB), OBJF(NB), OBJGRD(LDOBJD,NB), C(LDC,NB), CJAC(LDCJAC,SDCJAC,NB), R(LDR,SDR,NB), CLAMDA(LDCLDA,NB), OPTS(485), RUSER(*) 
LOGICAL 
REPEAT 
EXTERNAL 
CONFUN, OBJFUN, START 

Before calling E05UCF, the optional parameter arrays
IOPTS and
OPTS
must
be initialized for use with E05UCF by calling
E05ZKF with
OPTSTR set to
‘
Initialize = e05ucf’.
Optional parameters may be specified by calling
E05ZKF before the call to E05UCF.
3 Description
The problem is assumed to be stated in the following form:
where
$F\left(x\right)$ (the
objective function) is a nonlinear function,
${A}_{L}$ is an
${n}_{L}$ by
$n$ linear constraint matrix, and
$c\left(x\right)$ is an
${n}_{N}$ element vector of nonlinear constraint functions. (The matrix
${A}_{L}$ and the vector
$c\left(x\right)$ may be empty.) The objective function and the constraint functions are assumed to be smooth, i.e., at least twicecontinuously differentiable. (This routine will usually solve
(1) if there are only isolated discontinuities away from the solution.)
E05UCF solves a userspecified number of local optimization problems with different starting points. You may specify the starting points via the subroutine
START. If a random number generator is used to generate the starting points then the parameter
REPEAT allows you to specify whether a repeatable set of points are generated or whether different starting points are generated on different calls. The resulting local minima are ordered and the best
NB results returned in order of ascending values of the resulting objective function values at the minima. Thus the value returned in position
$1$ will be the best result obtained. If a sufficient number of different points are chosen then this is likely to be be the global minimum. Please note that the default version of
START uses a random number generator to generate the starting points.
4 References
Dennis J E Jr and Moré J J (1977) QuasiNewton methods, motivation and theory SIAM Rev. 19 46–89
Dennis J E Jr and Schnabel R B (1981) A new derivation of symmetric positivedefinite secant updates nonlinear programming (eds O L Mangasarian, R R Meyer and S M Robinson) 4 167–199 Academic Press
Dennis J E Jr and Schnabel R B (1983) Numerical Methods for Unconstrained Optimization and Nonlinear Equations Prentice–Hall
Fletcher R (1987) Practical Methods of Optimization (2nd Edition) Wiley
Gill P E, Hammarling S, Murray W, Saunders M A and Wright M H (1986) Users' guide for LSSOL (Version 1.0) Report SOL 861 Department of Operations Research, Stanford University
Gill P E, Murray W, Saunders M A and Wright M H (1984) Users' guide for SOL/QPSOL version 3.2 Report SOL 84–5 Department of Operations Research, Stanford University
Gill P E, Murray W, Saunders M A and Wright M H (1986a) Some theoretical properties of an augmented Lagrangian merit function Report SOL 86–6R Department of Operations Research, Stanford University
Gill P E, Murray W, Saunders M A and Wright M H (1986b) Users' guide for NPSOL (Version 4.0): a Fortran package for nonlinear programming Report SOL 862 Department of Operations Research, Stanford University
Gill P E, Murray W and Wright M H (1981) Practical Optimization Academic Press
Powell M J D (1974) Introduction to constrained optimization Numerical Methods for Constrained Optimization (eds P E Gill and W Murray) 1–28 Academic Press
Powell M J D (1983) Variable metric methods in constrained optimization Mathematical Programming: the State of the Art (eds A Bachem, M Grötschel and B Korte) 288–311 Springer–Verlag
5 Parameters
 1: $\mathrm{N}$ – INTEGERInput

On entry: $n$, the number of variables.
Constraint:
${\mathbf{N}}>0$.
 2: $\mathrm{NCLIN}$ – INTEGERInput

On entry: ${n}_{L}$, the number of general linear constraints.
Constraint:
${\mathbf{NCLIN}}\ge 0$.
 3: $\mathrm{NCNLN}$ – INTEGERInput

On entry: ${n}_{N}$, the number of nonlinear constraints.
Constraint:
${\mathbf{NCNLN}}\ge 0$.
 4: $\mathrm{A}\left({\mathbf{LDA}},*\right)$ – REAL (KIND=nag_wp) arrayInput

Note: the second dimension of the array
A
must be at least
${\mathbf{N}}$ if
${\mathbf{NCLIN}}>0$, and at least
$1$ otherwise.
On entry: the matrix
${A}_{L}$ of general linear constraints in
(1). That is, the
$\mathit{i}$th row contains the coefficients of the
$\mathit{i}$th general linear constraint, for
$\mathit{i}=1,2,\dots ,{\mathbf{NCLIN}}$.
If
${\mathbf{NCLIN}}=0$, the array
A is not referenced.
 5: $\mathrm{LDA}$ – INTEGERInput

On entry: the first dimension of the array
A as declared in the (sub)program from which E05UCF is called.
Constraint:
${\mathbf{LDA}}\ge {\mathbf{NCLIN}}$.
 6: $\mathrm{BL}\left({\mathbf{N}}+{\mathbf{NCLIN}}+{\mathbf{NCNLN}}\right)$ – REAL (KIND=nag_wp) arrayInput
 7: $\mathrm{BU}\left({\mathbf{N}}+{\mathbf{NCLIN}}+{\mathbf{NCNLN}}\right)$ – REAL (KIND=nag_wp) arrayInput

On entry:
BL must contain the lower bounds and
BU the upper bounds for all the constraints in the following order. The first
$n$ elements of each array must contain the bounds on the variables, the next
${n}_{L}$ elements the bounds for the general linear constraints (if any) and the next
${n}_{N}$ elements the bounds for the general nonlinear constraints (if any). To specify a nonexistent lower bound (i.e.,
${l}_{j}=\infty $), set
${\mathbf{BL}}\left(j\right)\le \mathit{bigbnd}$, and to specify a nonexistent upper bound (i.e.,
${u}_{j}=+\infty $), set
${\mathbf{BU}}\left(j\right)\ge \mathit{bigbnd}$; the default value of
$\mathit{bigbnd}$ is
${10}^{20}$, but this may be changed by the optional parameter
Infinite Bound Size. To specify the
$j$th constraint as an equality, set
${\mathbf{BL}}\left(j\right)={\mathbf{BU}}\left(j\right)=\beta $, say, where
$\left\beta \right<\mathit{bigbnd}$.
Constraints:
 ${\mathbf{BL}}\left(\mathit{j}\right)\le {\mathbf{BU}}\left(\mathit{j}\right)$, for $\mathit{j}=1,2,\dots ,{\mathbf{N}}+{\mathbf{NCLIN}}+{\mathbf{NCNLN}}$;
 if ${\mathbf{BL}}\left(j\right)={\mathbf{BU}}\left(j\right)=\beta $, $\left\beta \right<\mathit{bigbnd}$.
 8: $\mathrm{CONFUN}$ – SUBROUTINE, supplied by the NAG Library or the user.External Procedure

CONFUN must calculate the vector
$c\left(x\right)$ of nonlinear constraint functions and (optionally) its Jacobian (
$\text{}=\frac{\partial c}{\partial x}$) for a specified
$n$element vector
$x$.
If there are no nonlinear constraints (i.e.,
${\mathbf{NCNLN}}=0$),
CONFUN will never be called by E05UCF and
CONFUN may be the dummy routine E04UDM.
(E04UDM is included in the NAG Library.)
If there are nonlinear constraints, the first call to
CONFUN will occur before the first call to
OBJFUN.
The specification of
CONFUN is:
SUBROUTINE CONFUN ( 
MODE, NCNLN, N, LDCJSL, NEEDC, X, C, CJSL, NSTATE, IUSER, RUSER) 
INTEGER 
MODE, NCNLN, N, LDCJSL, NEEDC(NCNLN), NSTATE, IUSER(*) 
REAL (KIND=nag_wp) 
X(N), C(NCNLN), CJSL(LDCJSL,N), RUSER(*) 

 1: $\mathrm{MODE}$ – INTEGERInput/Output

On entry: indicates which values must be assigned during each call of
CONFUN. Only the following values need be assigned, for each value of
$i$ such that
${\mathbf{NEEDC}}\left(i\right)>0$:
 ${\mathbf{MODE}}=0$
 ${\mathbf{C}}\left(i\right)$.
 ${\mathbf{MODE}}=1$
 All available elements in the $i$th row of CJSL.
 ${\mathbf{MODE}}=2$
 ${\mathbf{C}}\left(i\right)$ and all available elements in the $i$th row of CJSL.
On exit: may be set to a negative value if you wish to abandon the solution to the current local minimization problem. In this case E05UCF will move to the next local minimization problem.
 2: $\mathrm{NCNLN}$ – INTEGERInput

On entry: ${n}_{N}$, the number of nonlinear constraints.
 3: $\mathrm{N}$ – INTEGERInput

On entry: $n$, the number of variables.
 4: $\mathrm{LDCJSL}$ – INTEGERInput

On entry:
LDCJSL is the same value as
LDCJAC in the call to E05UCF.
 5: $\mathrm{NEEDC}\left({\mathbf{NCNLN}}\right)$ – INTEGER arrayInput

On entry: the indices of the elements of
C and/or
CJSL that must be evaluated by
CONFUN. If
${\mathbf{NEEDC}}\left(i\right)>0$,
${\mathbf{C}}\left(i\right)$ and/or the available elements of the
$i$th row of
CJSL (see parameter
MODE) must be evaluated at
$x$.
 6: $\mathrm{X}\left({\mathbf{N}}\right)$ – REAL (KIND=nag_wp) arrayInput

On entry: $x$, the vector of variables at which the constraint functions and/or the available elements of the constraint Jacobian are to be evaluated.
 7: $\mathrm{C}\left({\mathbf{NCNLN}}\right)$ – REAL (KIND=nag_wp) arrayOutput

On exit: if
${\mathbf{NEEDC}}\left(k\right)>0$ and
${\mathbf{MODE}}=0$ or
$2$,
${\mathbf{C}}\left(k\right)$ must contain the value of
${c}_{k}\left(x\right)$. The remaining elements of
C, corresponding to the nonpositive elements of
NEEDC, need not be set.
 8: $\mathrm{CJSL}\left({\mathbf{LDCJSL}},{\mathbf{N}}\right)$ – REAL (KIND=nag_wp) arrayInput/Output

CJSL may be regarded as a twodimensional ‘slice’ of the threedimensional array
CJAC of E05UCF.
On entry: unless
${\mathbf{Derivative\; Level}}=2$ or
$3$ (the default setting is
${\mathbf{Derivative\; Level}}=3$,
the elements of
CJSL are set to special values which enable E05UCF to detect whether they are changed by
CONFUN.
On exit: if
${\mathbf{NEEDC}}\left(k\right)>0$ and
${\mathbf{MODE}}=1$ or
$2$, the
$k$th row of
CJSL must contain the available elements of the vector
$\nabla {c}_{k}$ given by
where
$\frac{\partial {c}_{k}}{\partial {x}_{j}}$ is the partial derivative of the
$k$th constraint with respect to the
$j$th variable, evaluated at the point
$x$. See also the parameter
NSTATE. The remaining rows of
CJSL, corresponding to nonpositive elements of
NEEDC, need not be set.
If all elements of the constraint Jacobian are known (i.e.,
${\mathbf{Derivative\; Level}}=2$ or
$3$), any constant elements may be assigned to
CJSL one time only at the start of each local optimization. An element of
CJSL that is not subsequently assigned in
CONFUN will retain its initial value throughout the local optimization. Constant elements may be loaded into
CJSL during the first call to
CONFUN for the local optimization (signalled by the value
${\mathbf{NSTATE}}=1$). The ability to preload constants is useful when many Jacobian elements are identically zero, in which case
CJSL may be initialized to zero and nonzero elements may be reset by
CONFUN.
Note that constant nonzero elements do affect the values of the constraints. Thus, if
${\mathbf{CJSL}}\left(k,j\right)$ is set to a constant value, it need not be reset in subsequent calls to
CONFUN, but the value
${\mathbf{CJSL}}\left(k,j\right)\times {\mathbf{X}}\left(j\right)$ must nonetheless be added to
${\mathbf{C}}\left(k\right)$. For example, if
${\mathbf{CJSL}}\left(1,1\right)=2$ and
${\mathbf{CJSL}}\left(1,2\right)=5$ then the term
$2\times {\mathbf{X}}\left(1\right)5\times {\mathbf{X}}\left(2\right)$ must be included in the definition of
${\mathbf{C}}\left(1\right)$.
It must be emphasized that, if
${\mathbf{Derivative\; Level}}=0$ or
$1$, unassigned elements of
CJSL are not treated as constant; they are estimated by finite differences, at nontrivial expense. If you do not supply a value for the optional parameter
Difference Interval, an interval for each element of
$x$ is computed automatically at the start of each local optimization. The automatic procedure can usually identify constant elements of
CJSL, which are then computed once only by finite differences.
 9: $\mathrm{NSTATE}$ – INTEGERInput

On entry: if
${\mathbf{NSTATE}}=1$ then E05UCF is calling
CONFUN for the first time on the current local optimization problem. This parameter setting allows you to save computation time if certain data must be calculated only once.
 10: $\mathrm{IUSER}\left(*\right)$ – INTEGER arrayUser Workspace
 11: $\mathrm{RUSER}\left(*\right)$ – REAL (KIND=nag_wp) arrayUser Workspace

CONFUN is called with the parameters
IUSER and
RUSER as supplied to E05UCF. You are free to use the arrays
IUSER and
RUSER to supply information to
CONFUN as an alternative to using COMMON global variables.
CONFUN must either be a module subprogram USEd by, or declared as EXTERNAL in, the (sub)program from which E05UCF is called. Parameters denoted as
Input must
not be changed by this procedure.
CONFUN should be tested separately before being used in conjunction with E05UCF. See also the description of the optional parameter
Verify.
 9: $\mathrm{OBJFUN}$ – SUBROUTINE, supplied by the user.External Procedure

OBJFUN must calculate the objective function
$F\left(x\right)$ and (optionally) its gradient
$g\left(x\right)=\frac{\partial F}{\partial x}$ for a specified
$n$vector
$x$.
The specification of
OBJFUN is:
INTEGER 
MODE, N, NSTATE, IUSER(*) 
REAL (KIND=nag_wp) 
X(N), OBJF, OBJGRD(N), RUSER(*) 

 1: $\mathrm{MODE}$ – INTEGERInput/Output

On entry: indicates which values must be assigned during each call of
OBJFUN. Only the following values need be assigned:
 ${\mathbf{MODE}}=0$
 OBJF.
 ${\mathbf{MODE}}=1$
 All available elements of OBJGRD.
 ${\mathbf{MODE}}=2$
 OBJF and all available elements of OBJGRD.
On exit: may be set to a negative value if you wish to abandon the solution to the current local minimization problem. In this case E05UCF will move to the next local minimization problem.
 2: $\mathrm{N}$ – INTEGERInput

On entry: $n$, the number of variables.
 3: $\mathrm{X}\left({\mathbf{N}}\right)$ – REAL (KIND=nag_wp) arrayInput

On entry: $x$, the vector of variables at which the objective function and/or all available elements of its gradient are to be evaluated.
 4: $\mathrm{OBJF}$ – REAL (KIND=nag_wp)Output

On exit: if
${\mathbf{MODE}}=0$ or
$2$,
OBJF must be set to the value of the objective function at
$x$.
 5: $\mathrm{OBJGRD}\left({\mathbf{N}}\right)$ – REAL (KIND=nag_wp) arrayInput/Output

On entry: the elements of
OBJGRD are set to special values which enable E05UCF to detect whether they are changed by
OBJFUN.
On exit: if
${\mathbf{MODE}}=1$ or
$2$,
OBJGRD must return the available elements of the gradient evaluated at
$x$.
 6: $\mathrm{NSTATE}$ – INTEGERInput

On entry: if
${\mathbf{NSTATE}}=1$ then E05UCF is calling
OBJFUN for the first time on the current local optimization problem. This parameter setting allows you to save computation time if certain data must be calculated only once.
 7: $\mathrm{IUSER}\left(*\right)$ – INTEGER arrayUser Workspace
 8: $\mathrm{RUSER}\left(*\right)$ – REAL (KIND=nag_wp) arrayUser Workspace

OBJFUN is called with the parameters
IUSER and
RUSER as supplied to E05UCF. You are free to use the arrays
IUSER and
RUSER to supply information to
OBJFUN as an alternative to using COMMON global variables.
OBJFUN must either be a module subprogram USEd by, or declared as EXTERNAL in, the (sub)program from which E05UCF is called. Parameters denoted as
Input must
not be changed by this procedure.
OBJFUN should be tested separately before being used in conjunction with E05UCF. See also the description of the optional parameter
Verify.
 10: $\mathrm{NPTS}$ – INTEGERInput

On entry: the number of different starting points to be generated and used. The more points used, the more likely that the best returned solution will be a global minimum.
Constraint:
$1\le {\mathbf{NB}}\le {\mathbf{NPTS}}$.
 11: $\mathrm{X}\left({\mathbf{LDX}},{\mathbf{NB}}\right)$ – REAL (KIND=nag_wp) arrayOutput

On exit: ${\mathbf{X}}\left(\mathit{j},i\right)$ contains the final estimate of the $i$th solution, for $\mathit{j}=1,2,\dots ,{\mathbf{N}}$.
 12: $\mathrm{LDX}$ – INTEGERInput

On entry: the first dimension of the array
X as declared in the (sub)program from which E05UCF is called.
Constraint:
${\mathbf{LDX}}\ge {\mathbf{N}}$.
 13: $\mathrm{START}$ – SUBROUTINE, supplied by the NAG Library or the user.External Procedure

START must calculate the
NPTS starting points to be used by the local optimizer. If you do not wish to write a routine specific to your problem then E05UCZ may be used as the actual argument. E05UCZ is supplied in the NAG Library and uses the NAG quasirandom number generators to distribute starting points uniformly across the domain. It is affected by the value of
REPEAT.
The specification of
START is:
INTEGER 
NPTS, N, IUSER(*), MODE 
REAL (KIND=nag_wp) 
QUAS(N,NPTS), BL(N), BU(N), RUSER(*) 
LOGICAL 
REPEAT 

 1: $\mathrm{NPTS}$ – INTEGERInput

On entry: indicates the number of starting points.
 2: $\mathrm{QUAS}\left({\mathbf{N}},{\mathbf{NPTS}}\right)$ – REAL (KIND=nag_wp) arrayInput/Output

On entry: all elements of
QUAS will have been set to zero, so only nonzero values need be set subsequently.
On exit: must contain the starting points for the
NPTS local minimizations, i.e.,
${\mathbf{QUAS}}\left(j,i\right)$ must contain the
$j$th component of the
$i$th starting point.
 3: $\mathrm{N}$ – INTEGERInput

On entry: the number of variables.
 4: $\mathrm{REPEAT}$ – LOGICALInput

On entry: specifies whether a repeatable or nonrepeatable sequence of points are to be generated.
 5: $\mathrm{BL}\left({\mathbf{N}}\right)$ – REAL (KIND=nag_wp) arrayInput

On entry: the lower bounds on the variables. These may be used to ensure that the starting points generated in some sense ‘cover’ the region, but there is no requirement that a starting point be feasible.
 6: $\mathrm{BU}\left({\mathbf{N}}\right)$ – REAL (KIND=nag_wp) arrayInput

On entry: the upper bounds on the variables. (See
BL.)
 7: $\mathrm{IUSER}\left(*\right)$ – INTEGER arrayUser Workspace
 8: $\mathrm{RUSER}\left(*\right)$ – REAL (KIND=nag_wp) arrayUser Workspace

START is called with the parameters
IUSER and
RUSER as supplied to E05UCF. You are free to use the arrays
IUSER and
RUSER to supply information to
START as an alternative to using COMMON global variables.
 9: $\mathrm{MODE}$ – INTEGERInput/Output

On entry:
MODE will contain
$0$.
On exit: if you set
MODE to a negative value then E05UCF will terminate immediately with
${\mathbf{IFAIL}}={\mathbf{9}}$.
START must either be a module subprogram USEd by, or declared as EXTERNAL in, the (sub)program from which E05UCF is called. Parameters denoted as
Input must
not be changed by this procedure.
 14: $\mathrm{REPEAT}$ – LOGICALInput

On entry: is passed as an argument to
START and may be used to initialize a random number generator to a repeatable, or nonrepeatable, sequence.
 15: $\mathrm{NB}$ – INTEGERInput

On entry: the number of solutions to be returned. The routine saves up to
NB local minima ordered by increasing value of the final objective function. If the defining criterion for ‘best solution’ is only that the value of the objective function is as small as possible then
NB should be set to
$1$. However, if you want to look at other solutions that may have desirable properties then setting
${\mathbf{NB}}>1$ will produce
NB local minima, ordered by increasing value of their objective functions at the minima.
Constraint:
$1\le {\mathbf{NB}}\le {\mathbf{NPTS}}$.
 16: $\mathrm{OBJF}\left({\mathbf{NB}}\right)$ – REAL (KIND=nag_wp) arrayOutput

On exit: ${\mathbf{OBJF}}\left(i\right)$ contains the value of the objective function at the final iterate for the $i$th solution.
 17: $\mathrm{OBJGRD}\left({\mathbf{LDOBJD}},{\mathbf{NB}}\right)$ – REAL (KIND=nag_wp) arrayOutput

On exit: ${\mathbf{OBJGRD}}\left(\mathit{j},i\right)$ contains the gradient of the objective function for the $i$th solution at the final iterate (or its finite difference approximation), for $\mathit{j}=1,2,\dots ,{\mathbf{N}}$.
 18: $\mathrm{LDOBJD}$ – INTEGERInput

On entry: the first dimension of the array
OBJGRD as declared in the (sub)program from which E05UCF is called.
Constraint:
${\mathbf{LDOBJD}}\ge {\mathbf{N}}$.
 19: $\mathrm{ITER}\left({\mathbf{NB}}\right)$ – INTEGER arrayOutput

On exit:
${\mathbf{ITER}}\left(i\right)$ contains the number of major iterations performed to obtain the
$i$th solution. If less than
NB solutions are returned then
${\mathbf{ITER}}\left({\mathbf{NB}}\right)$ contains the number of starting points that have resulted in a converged solution. If this is close to
NPTS then this might be indicative that fewer than
NB local minima exist.
 20: $\mathrm{C}\left({\mathbf{LDC}},{\mathbf{NB}}\right)$ – REAL (KIND=nag_wp) arrayOutput

On exit: if
${\mathbf{NCNLN}}>0$,
${\mathbf{C}}\left(\mathit{j},\mathit{i}\right)$ contains the value of the
$\mathit{j}$th nonlinear constraint function
${c}_{\mathit{j}}$ at the final iterate, for the
$\mathit{i}$th solution, for
$\mathit{j}=1,2,\dots ,{\mathbf{NCNLN}}$.
If
${\mathbf{NCNLN}}=0$, the array
C is not referenced.
 21: $\mathrm{LDC}$ – INTEGERInput

On entry: the first dimension of the array
C as declared in the (sub)program from which E05UCF is called.
Constraint:
${\mathbf{LDC}}\ge {\mathbf{NCNLN}}$.
 22: $\mathrm{CJAC}\left({\mathbf{LDCJAC}},{\mathbf{SDCJAC}},{\mathbf{NB}}\right)$ – REAL (KIND=nag_wp) arrayOutput

On exit: if
${\mathbf{NCNLN}}>0$,
CJAC contains the Jacobian matrices of the nonlinear constraint functions at the final iterate for each of the returned solutions, i.e.,
${\mathbf{CJAC}}\left(\mathit{k},\mathit{j},i\right)$ contains the partial derivative of the
$\mathit{k}$th constraint function with respect to the
$\mathit{j}$th variable, for
$\mathit{k}=1,2,\dots ,{\mathbf{NCNLN}}$ and
$\mathit{j}=1,2,\dots ,{\mathbf{N}}$, for the
$i$th solution. (See the discussion of parameter
CJSL under
CONFUN.)
If
${\mathbf{NCNLN}}=0$, the array
CJAC is not referenced.
 23: $\mathrm{LDCJAC}$ – INTEGERInput

On entry: the first dimension of the array
CJAC as declared in the (sub)program from which E05UCF is called.
Constraint:
${\mathbf{LDCJAC}}\ge {\mathbf{NCNLN}}$.
 24: $\mathrm{SDCJAC}$ – INTEGERInput

On entry: the second dimension of the array
CJAC as declared in the (sub)program from which E05UCF is called.
Constraint:
if ${\mathbf{NCNLN}}>0$, ${\mathbf{SDCJAC}}\ge {\mathbf{N}}$.
 25: $\mathrm{R}\left({\mathbf{LDR}},{\mathbf{SDR}},{\mathbf{NB}}\right)$ – REAL (KIND=nag_wp) arrayOutput

On exit: for each of the
NB solutions
R will contain a form of the Hessian; for the
$i$th returned solution
${\mathbf{R}}\left({\mathbf{LDR}},{\mathbf{SDR}},i\right)$ contains the Hessian that would be returned from the local minimizer. If
${\mathbf{Hessian}}=\mathrm{NO}$,
the default, each
${\mathbf{R}}\left({\mathbf{LDR}},{\mathbf{SDR}},i\right)$ contains the upper triangular Cholesky factor
$R$ of
${Q}^{\mathrm{T}}HQ$, an estimate of the transformed and reordered Hessian of the Lagrangian at
$x$. If
${\mathbf{Hessian}}=\mathrm{YES}$,
${\mathbf{R}}\left({\mathbf{LDR}},{\mathbf{SDR}},i\right)$ contains the upper triangular Cholesky factor
$R$ of
$H$, the approximate (untransformed) Hessian of the Lagrangian, with the variables in the natural order.
 26: $\mathrm{LDR}$ – INTEGERInput

On entry: the first dimension of the array
R as declared in the (sub)program from which E05UCF is called.
Constraint:
${\mathbf{LDR}}\ge {\mathbf{N}}$.
 27: $\mathrm{SDR}$ – INTEGERInput

On entry: the second dimension of the array
R as declared in the (sub)program from which E05UCF is called.
Constraint:
${\mathbf{SDR}}\ge {\mathbf{N}}$.
 28: $\mathrm{CLAMDA}\left({\mathbf{LDCLDA}},{\mathbf{NB}}\right)$ – REAL (KIND=nag_wp) arrayOutput

On exit: the values of the QP multipliers from the last QP subproblem solved for the $i$th solution. ${\mathbf{CLAMDA}}\left(j,i\right)$ should be nonnegative if ${\mathbf{ISTATE}}\left(j,i\right)=1$ and nonpositive if ${\mathbf{ISTATE}}\left(j,i\right)=2$.
 29: $\mathrm{LDCLDA}$ – INTEGERInput

On entry: the first dimension of the array
CLAMDA as declared in the (sub)program from which E05UCF is called.
Constraint:
${\mathbf{LDCLDA}}\ge {\mathbf{N}}+{\mathbf{NCLIN}}+{\mathbf{NCNLN}}$.
 30: $\mathrm{ISTATE}\left({\mathbf{LISTAT}},{\mathbf{NB}}\right)$ – INTEGER arrayOutput

On exit:
${\mathbf{ISTATE}}\left(j,i\right)$ contains the status of the constraints in the QP working set for the
$i$th solution. The significance of each possible value of
${\mathbf{ISTATE}}\left(j,i\right)$ is as follows:
${\mathbf{ISTATE}}\left(j,i\right)$  Meaning 
$\phantom{}0$  The constraint is satisfied to within the feasibility tolerance, but is not in the QP working set. 
$\phantom{}1$  This inequality constraint is included in the QP working set at its lower bound. 
$\phantom{}2$  This inequality constraint is included in the QP working set at its upper bound. 
$\phantom{}3$  This constraint is included in the QP working set as an equality. This value of ISTATE can occur only when ${\mathbf{BL}}\left(j\right)={\mathbf{BU}}\left(j\right)$. 
 31: $\mathrm{LISTAT}$ – INTEGERInput

On entry: the first dimension of the array
ISTATE as declared in the (sub)program from which E05UCF is called.
Constraint:
${\mathbf{LISTAT}}\ge {\mathbf{N}}+{\mathbf{NCLIN}}+{\mathbf{NCNLN}}$.
 32: $\mathrm{IOPTS}\left(740\right)$ – INTEGER arrayCommunication Array
 33: $\mathrm{OPTS}\left(485\right)$ – REAL (KIND=nag_wp) arrayCommunication Array

The arrays
IOPTS and
OPTS must not be altered between calls to any of the routines E05UCF and
E05ZKF.
 34: $\mathrm{IUSER}\left(*\right)$ – INTEGER arrayUser Workspace
 35: $\mathrm{RUSER}\left(*\right)$ – REAL (KIND=nag_wp) arrayUser Workspace

IUSER and
RUSER are not used by E05UCF, but are passed directly to
CONFUN,
OBJFUN and
START and may be used to pass information to these routines as an alternative to using COMMON global variables.
With care, you may also write information back into
IUSER and
RUSER. This might be useful, for example, should there be a need to preserve the state of a random number generator.
With SMPenabled versions of E05UCF the arrays
IUSER and
RUSER provided are classified as OpenMP shared memory. Use of
IUSER and
RUSER has to take account of this in order to preserve thread safety whenever information is written back to either of these arrays.
 36: $\mathrm{INFO}\left({\mathbf{NB}}\right)$ – INTEGER arrayOutput

On exit:
${\mathbf{INFO}}\left(i\right)$ contains one of
$0$,
$1$ or
$6$. Please see the description of each corresponding value of
IFAIL on exit from
E04UCF/E04UCA for detailed explanations of these exit values. As usual
$0$ denotes success.
If
${\mathbf{IFAIL}}={\mathbf{8}}$ on exit, then not all
NB solutions have been found, and
${\mathbf{INFO}}\left({\mathbf{NB}}\right)$ contains the number of solutions actually found.
 37: $\mathrm{IFAIL}$ – INTEGERInput/Output

On entry:
IFAIL must be set to
$0$,
$1\text{ or}1$. If you are unfamiliar with this parameter you should refer to
Section 3.3 in the Essential Introduction for details.
For environments where it might be inappropriate to halt program execution when an error is detected, the value
$1\text{ or}1$ is recommended. If the output of error messages is undesirable, then the value
$1$ is recommended. Otherwise, because for this routine the values of the output parameters may be useful even if
${\mathbf{IFAIL}}\ne {\mathbf{0}}$ on exit, the recommended value is
$1$.
When the value $\mathbf{1}\text{ or}1$ is used it is essential to test the value of IFAIL on exit.
On exit:
${\mathbf{IFAIL}}={\mathbf{0}}$ unless the routine detects an error or a warning has been flagged (see
Section 6).
6 Error Indicators and Warnings
If on entry
${\mathbf{IFAIL}}={\mathbf{0}}$ or
${{\mathbf{1}}}$, explanatory error messages are output on the current error message unit (as defined by
X04AAF).
Note: E05UCF may return useful information for one or more of the following detected errors or warnings.
Errors or warnings detected by the routine:
 ${\mathbf{IFAIL}}=1$
An input value is incorrect. One or more of the following constraints are violated.

On entry, ${\mathbf{BL}}\left(i\right)>{\mathbf{BU}}\left(i\right)$: $i=\u2329\mathit{\text{value}}\u232a$.
Constraint: ${\mathbf{BL}}\left(i\right)\le {\mathbf{BU}}\left(i\right)$, for all $i$.

On entry, ${\mathbf{LDA}}=\u2329\mathit{\text{value}}\u232a$ and ${\mathbf{NCLIN}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: ${\mathbf{LDA}}\ge {\mathbf{NCLIN}}$.

On entry, ${\mathbf{LDC}}=\u2329\mathit{\text{value}}\u232a$ and ${\mathbf{NCNLN}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: ${\mathbf{LDC}}\ge {\mathbf{NCNLN}}$.

On entry, ${\mathbf{LDCJAC}}=\u2329\mathit{\text{value}}\u232a$ and ${\mathbf{NCNLN}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: ${\mathbf{LDCJAC}}\ge {\mathbf{NCNLN}}$.

On entry, ${\mathbf{LDCLDA}}=\u2329\mathit{\text{value}}\u232a$, ${\mathbf{N}}=\u2329\mathit{\text{value}}\u232a$, ${\mathbf{NCLIN}}=\u2329\mathit{\text{value}}\u232a$ and ${\mathbf{NCNLN}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: ${\mathbf{LDCLDA}}\ge {\mathbf{N}}+{\mathbf{NCLIN}}+{\mathbf{NCNLN}}$.

On entry, ${\mathbf{LDOBJD}}=\u2329\mathit{\text{value}}\u232a$ and ${\mathbf{N}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: ${\mathbf{LDOBJD}}\ge {\mathbf{N}}$.

On entry, ${\mathbf{LDR}}=\u2329\mathit{\text{value}}\u232a$ and ${\mathbf{N}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: ${\mathbf{LDR}}\ge {\mathbf{N}}$.

On entry, ${\mathbf{LDX}}=\u2329\mathit{\text{value}}\u232a$ and ${\mathbf{N}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: ${\mathbf{LDX}}\ge {\mathbf{N}}$.

On entry, ${\mathbf{LISTAT}}=\u2329\mathit{\text{value}}\u232a$, ${\mathbf{N}}=\u2329\mathit{\text{value}}\u232a$, ${\mathbf{NCLIN}}=\u2329\mathit{\text{value}}\u232a$ and ${\mathbf{NCNLN}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: ${\mathbf{LISTAT}}\ge {\mathbf{N}}+{\mathbf{NCLIN}}+{\mathbf{NCNLN}}$.

On entry, ${\mathbf{N}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: ${\mathbf{N}}>0$.

On entry, ${\mathbf{NB}}=\u2329\mathit{\text{value}}\u232a$ and ${\mathbf{NPTS}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: $1\le {\mathbf{NB}}\le {\mathbf{NPTS}}$.

On entry, ${\mathbf{NCLIN}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: ${\mathbf{NCLIN}}\ge 0$.

On entry, ${\mathbf{NCNLN}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: ${\mathbf{NCNLN}}\ge 0$.

On entry, ${\mathbf{NCNLN}}>0$, ${\mathbf{SDCJAC}}=\u2329\mathit{\text{value}}\u232a$ and ${\mathbf{N}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: if ${\mathbf{NCNLN}}>0$, ${\mathbf{SDCJAC}}\ge {\mathbf{N}}$.

On entry, ${\mathbf{SDR}}=\u2329\mathit{\text{value}}\u232a$ and ${\mathbf{N}}=\u2329\mathit{\text{value}}\u232a$.
Constraint: ${\mathbf{SDR}}\ge {\mathbf{N}}$.
 ${\mathbf{IFAIL}}=2$

No solution obtained. Linear constraints may be infeasible.
E05UCF has terminated without finding any solutions. The majority of calls to the local optimizer have failed to find a feasible point for the linear constraints and bounds, which means that either no feasible point exists for the given value of the optional parameter Linear Feasibility Tolerance (default value $\sqrt{\epsilon}$, where $\epsilon $ is the machine precision), or no feasible point could be found in the number of iterations specified by the optional parameter Minor Iteration Limit. You should check that there are no constraint redundancies. If the data for the constraints are accurate only to an absolute precision $\sigma $, you should ensure that the value of the optional parameter Linear Feasibility Tolerance is greater than $\sigma $. For example, if all elements of ${A}_{L}$ are of order unity and are accurate to only three decimal places, Linear Feasibility Tolerance should be at least ${10}^{3}$.
 ${\mathbf{IFAIL}}=3$

E05UCF has failed to find any solutions. The majority of local optimizations could not find a feasible point for the nonlinear constraints. The problem may have no feasible solution. This behaviour will occur if there is no feasible point for the nonlinear constraints. (However, there is no general test that can determine whether a feasible point exists for a set of nonlinear constraints.)
No solution obtained. Nonlinear constraints may be infeasible.
 ${\mathbf{IFAIL}}=4$

No solution obtained. Many potential solutions reach iteration limit.
The Iteration Limit may be changed using E05ZKF.
 ${\mathbf{IFAIL}}=7$

Usersupplied derivatives probably wrong.
The usersupplied derivatives of the objective function and/or nonlinear constraints appear to be incorrect.
Large errors were found in the derivatives of the objective function and/or nonlinear constraints. This value of
IFAIL will occur if the verification process indicated that at least one gradient or Jacobian element had no correct figures. You should refer to or enable the printed output to determine which elements are suspected to be in error.
As a firststep, you should check that the code for the objective and constraint values is correct – for example, by computing the function at a point where the correct value is known. However, care should be taken that the chosen point fully tests the evaluation of the function. It is remarkable how often the values $x=0$ or $x=1$ are used to test function evaluation procedures, and how often the special properties of these numbers make the test meaningless.
Gradient checking will be ineffective if the objective function uses information computed by the constraints, since they are not necessarily computed before each function evaluation.
Errors in programming the function may be quite subtle in that the function value is ‘almost’ correct. For example, the function may not be accurate to full precision because of the inaccurate calculation of a subsidiary quantity, or the limited accuracy of data upon which the function depends. A common error on machines where numerical calculations are usually performed in double precision is to include even one single precision constant in the calculation of the function; since some compilers do not convert such constants to double precision, half the correct figures may be lost by such a seemingly trivial error.
 ${\mathbf{IFAIL}}=8$

Only
$\u2329\mathit{\text{value}}\u232a$ solutions obtained.
Not all
NB solutions have been found.
${\mathbf{INFO}}\left({\mathbf{NB}}\right)$ contains the number actually found.
 ${\mathbf{IFAIL}}=9$

User terminated computation from
START procedure:
${\mathbf{MODE}}=\u2329\mathit{\text{value}}\u232a$.
If
E05UCZ
has been used as an actual argument for START then the message displayed, when ${\mathbf{IFAIL}}={\mathbf{0}}$ or ${{\mathbf{1}}}$ on entry to E05UCF, will have the following meaning:
$998$ 
failure to allocate space, a smaller value of NPTS should be tried. 
$997$ 
an internal error has occurred. Please contact NAG for assistance. 
 ${\mathbf{IFAIL}}=10$

Failed to initialize optional parameter arrays.
 ${\mathbf{IFAIL}}=99$
An unexpected error has been triggered by this routine. Please
contact
NAG.
See
Section 3.8 in the Essential Introduction for further information.
 ${\mathbf{IFAIL}}=399$
Your licence key may have expired or may not have been installed correctly.
See
Section 3.7 in the Essential Introduction for further information.
 ${\mathbf{IFAIL}}=999$
Dynamic memory allocation failed.
See
Section 3.6 in the Essential Introduction for further information.
7 Accuracy
If
${\mathbf{IFAIL}}={\mathbf{0}}$ on exit and the value of
${\mathbf{INFO}}\left(i\right)=0$, then the vector returned in the array
X for solution
$i$ is an estimate of the solution to an accuracy of approximately
Optimality Tolerance.
8 Parallelism and Performance
E05UCF is threaded by NAG for parallel execution in multithreaded implementations of the NAG Library. In these implementations, this routine may make calls to the usersupplied functions from within an OpenMP parallel region. Thus OpenMP directives within the user functions can only be used if you are compiling the usersupplied function and linking the executable in accordance with the instructions in the
Users' Note for your implementation. The user workspace arrays
IUSER and
RUSER are classified as OpenMP shared memory and use of
IUSER and
RUSER has to take account of this in order to preserve thread safety whenever information is written back to either of these arrays. If at all possible, it is recommended that these arrays are only used to supply readonly data to the user functions when a multithreaded implementation is being used.
E05UCF makes calls to BLAS and/or LAPACK routines, which may be threaded within the vendor library used by this implementation. Consult the documentation for the vendor library for further information.
Please consult the
X06 Chapter Introduction for information on how to control and interrogate the OpenMP environment used within this routine. Please also consult the
Users' Note for your implementation for any additional implementationspecific information.
You should be wary of requesting much intermediate output from the local optimizer, since large volumes may be produced if
NPTS is large.
The auxiliary routine E05UCZ
makes use of the NAG quasirandom Sobol generator (
G05YLF and
G05YMF). If
E05UCZ
is used as an argument for
START (see the description of
START) and
${\mathbf{REPEAT}}=\mathrm{.FALSE.}$ then a randomly chosen value for
ISKIP is used, otherwise
ISKIP is set to
$100$. If
REPEAT is set to .FALSE. and the program is executed several times, each time producing the same best answer, then there is increased probability that this answer is a global minimum. However, if it is important that identical results be obtained on successive runs, then
REPEAT should be set to .TRUE..
9.1 Description of the Printed Output
See
Section 9.1 in E04UCF/E04UCA.
10 Example
This example finds the global minimum of the twodimensional Schwefel function:
subject to the constraints:
10.1 Program Text
Program Text (e05ucfe.f90)
10.2 Program Data
Program Data (e05ucfe.d)
10.3 Program Results
Program Results (e05ucfe.r)
11 Algorithmic Details
See
Section 11 in E04UCF/E04UCA.
12 Optional Parameters
Several optional parameters in E05UCF define choices in the problem specification or the algorithm logic. In order to reduce the number of formal parameters of E05UCF these optional parameters have associated default values that are appropriate for most problems. Therefore you need only specify those optional parameters whose values are to be different from their default values.
Optional parameters may be specified by calling
E05ZKF before a call to E05UCF. Before calling E05UCF, the optional parameter arrays
IOPTS and
OPTS
must
be initialized for use with E05UCF by calling
E05ZKF with
OPTSTR set to ‘
Initialize = e05ucf’.
All optional parameters not specified are set to their default values. Optional parameters specified are unaltered by E05UCF (unless they define invalid values) and so remain in effect for subsequent calls to E05UCF.
12.1 Description of the Optional Parameters
E05UCF supports two options that are distinct from those of
E04UCF/E04UCA:
Punch Unit  $i$  Default $\text{}=6$ 
This option allows you to send information arising from an appropriate setting of
Out_Level to be sent to the Fortran unit number defined by
Punch Unit. If you wish this file to be different to the standard output unit (
$6$) where other output is displayed then this file should be attached by calling
X04ACF prior to calling E05UCF.
Out_Level  $i$  Default $\text{}=0$ 
This option defines the amount of extra information to be sent to the Fortran unit number defined by
Punch Unit. The possible choices for
$i$ are the following:
$i$ 
Meaning 
0 
No extra output. 
1 
Updated solutions only. This is useful during long runs to observe progress. 
2 
Successful start points only. This is useful to save the starting points that gave rise to the final solution. 
3 
Both updated solutions and successful start points. 
See
Section 12 in E04UCF/E04UCA for details of the other options.
The
Warm Start option of
E04UCF/E04UCA is not a valid option for use with E05UCF.
13 Description of Monitoring Information
See
Section 13 in E04UCF/E04UCA.