H Chapter Contents
H Chapter Introduction
NAG Library Manual

Note:  before using this routine, please read the Users' Note for your implementation to check the interpretation of bold italicised terms and other implementation-dependent details.

## 1  Purpose

H03ADF finds the shortest path through a directed or undirected acyclic network using Dijkstra's algorithm.

## 2  Specification

 SUBROUTINE H03ADF ( N, NS, NE, DIRECT, NNZ, D, IROW, ICOL, SPLEN, PATH, IWORK, WORK, IFAIL)
 INTEGER N, NS, NE, NNZ, IROW(NNZ), ICOL(NNZ), PATH(N), IWORK(3*N+1), IFAIL REAL (KIND=nag_wp) D(NNZ), SPLEN, WORK(2*N) LOGICAL DIRECT

## 3  Description

H03ADF attempts to find the shortest path through a directed or undirected acyclic network, which consists of a set of points called vertices and a set of curves called arcs that connect certain pairs of distinct vertices. An acyclic network is one in which there are no paths connecting a vertex to itself. An arc whose origin vertex is $i$ and whose destination vertex is $j$ can be written as $i\to j$. In an undirected network the arcs $i\to j$ and $j\to i$ are equivalent (i.e., $i↔j$), whereas in a directed network they are different. Note that the shortest path may not be unique and in some cases may not even exist (e.g., if the network is disconnected).
The network is assumed to consist of $n$ vertices which are labelled by the integers $1,2,\dots ,n$. The lengths of the arcs between the vertices are defined by the $n$ by $n$ distance matrix ${\mathbf{D}}$, in which the element ${d}_{ij}$ gives the length of the arc $i\to j$; ${d}_{ij}=0$ if there is no arc connecting vertices $i$ and $j$ (as is the case for an acyclic network when $i=j$). Thus the matrix $D$ is usually sparse. For example, if $n=4$ and the network is directed, then
 $D= 0 d12 d13 d14 d21 0 d23 d24 d31 d32 0 d34 d41 d42 d43 0 .$
If the network is undirected, ${\mathbf{D}}$ is symmetric since ${d}_{ij}={d}_{ji}$ (i.e., the length of the arc $i\to j\equiv \text{}$ the length of the arc $j\to i$).
The method used by H03ADF is described in detail in Section 8.

## 4  References

Dijkstra E W (1959) A note on two problems in connection with graphs Numer. Math. 1 269–271

## 5  Parameters

1:     N – INTEGERInput
On entry: $n$, the number of vertices.
Constraint: ${\mathbf{N}}\ge 2$.
2:     NS – INTEGERInput
3:     NE – INTEGERInput
On entry: ${n}_{s}$ and ${n}_{e}$, the labels of the first and last vertices, respectively, between which the shortest path is sought.
Constraints:
• $1\le {\mathbf{NS}}\le {\mathbf{N}}$;
• $1\le {\mathbf{NE}}\le {\mathbf{N}}$;
• ${\mathbf{NS}}\ne {\mathbf{NE}}$.
4:     DIRECT – LOGICALInput
On entry: indicates whether the network is directed or undirected.
${\mathbf{DIRECT}}=\mathrm{.TRUE.}$
The network is directed.
${\mathbf{DIRECT}}=\mathrm{.FALSE.}$
The network is undirected.
5:     NNZ – INTEGERInput
On entry: the number of nonzero elements in the distance matrix $D$.
Constraints:
• if ${\mathbf{DIRECT}}=\mathrm{.TRUE.}$, $1\le {\mathbf{NNZ}}\le {\mathbf{N}}×\left({\mathbf{N}}-1\right)$;
• if ${\mathbf{DIRECT}}=\mathrm{.FALSE.}$, $1\le {\mathbf{NNZ}}\le {\mathbf{N}}×\left({\mathbf{N}}-1\right)/2$.
6:     D(NNZ) – REAL (KIND=nag_wp) arrayInput
On entry: the nonzero elements of the distance matrix $D$, ordered by increasing row index and increasing column index within each row. More precisely, ${\mathbf{D}}\left(k\right)$ must contain the value of the nonzero element with indices (${\mathbf{IROW}}\left(k\right),{\mathbf{ICOL}}\left(k\right)$); this is the length of the arc from the vertex with label ${\mathbf{IROW}}\left(k\right)$ to the vertex with label ${\mathbf{ICOL}}\left(k\right)$. Elements with the same row and column indices are not allowed. If ${\mathbf{DIRECT}}=\mathrm{.FALSE.}$, then only those nonzero elements in the strict upper triangle of ${\mathbf{D}}$ need be supplied since ${d}_{ij}={d}_{ji}$. (F11ZAF may be used to sort the elements of an arbitrarily ordered matrix into the required form. This is illustrated in Section 9.)
Constraint: ${\mathbf{D}}\left(\mathit{k}\right)>0.0$, for $\mathit{k}=1,2,\dots ,{\mathbf{NNZ}}$.
7:     IROW(NNZ) – INTEGER arrayInput
8:     ICOL(NNZ) – INTEGER arrayInput
On entry: ${\mathbf{IROW}}\left(k\right)$ and ${\mathbf{ICOL}}\left(k\right)$ must contain the row and column indices, respectively, for the nonzero element stored in ${\mathbf{D}}\left(k\right)$.
Constraints:
IROW and ICOL must satisfy the following constraints (which may be imposed by a call to F11ZAF):
• ${\mathbf{IROW}}\left(k-1\right)<{\mathbf{IROW}}\left(k\right)$;
• ${\mathbf{IROW}}\left(\mathit{k}-1\right)={\mathbf{IROW}}\left(\mathit{k}\right)$ and ${\mathbf{ICOL}}\left(\mathit{k}-1\right)<{\mathbf{ICOL}}\left(\mathit{k}\right)$, for $\mathit{k}=2,3,\dots ,{\mathbf{NNZ}}$.
In addition, if ${\mathbf{DIRECT}}=\mathrm{.TRUE.}$, $1\le {\mathbf{IROW}}\left(k\right)\le {\mathbf{N}}$, $1\le {\mathbf{ICOL}}\left(k\right)\le {\mathbf{N}}$ and ${\mathbf{IROW}}\left(k\right)\ne {\mathbf{ICOL}}\left(k\right)$;
• if ${\mathbf{DIRECT}}=\mathrm{.FALSE.}$, $1\le {\mathbf{IROW}}\left(k\right)<{\mathbf{ICOL}}\left(k\right)\le {\mathbf{N}}$.
9:     SPLEN – REAL (KIND=nag_wp)Output
On exit: the length of the shortest path between the specified vertices ${n}_{s}$ and ${n}_{e}$.
10:   PATH(N) – INTEGER arrayOutput
On exit: contains details of the shortest path between the specified vertices ${n}_{s}$ and ${n}_{e}$. More precisely, ${\mathbf{NS}}={\mathbf{PATH}}\left(1\right)\to {\mathbf{PATH}}\left(2\right)\to \dots \to {\mathbf{PATH}}\left(p\right)={\mathbf{NE}}$ for some $p\le n$. The remaining $\left(n-p\right)$ elements are set to zero.
11:   IWORK($3×{\mathbf{N}}+1$) – INTEGER arrayWorkspace
12:   WORK($2×{\mathbf{N}}$) – REAL (KIND=nag_wp) arrayWorkspace
13:   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, if you are not familiar with this parameter, the recommended value is $0$. When the value $-\mathbf{1}\text{​ or ​}\mathbf{1}$ 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).
Errors or warnings detected by the routine:
${\mathbf{IFAIL}}=1$
 On entry, ${\mathbf{N}}<2$, or ${\mathbf{NS}}<1$, or ${\mathbf{NS}}>{\mathbf{N}}$, or ${\mathbf{NE}}<1$, or ${\mathbf{NE}}>{\mathbf{N}}$, or ${\mathbf{NS}}={\mathbf{NE}}$.
${\mathbf{IFAIL}}=2$
 On entry, ${\mathbf{NNZ}}>{\mathbf{N}}×\left({\mathbf{N}}-1\right)$ when ${\mathbf{DIRECT}}=\mathrm{.TRUE.}$, or ${\mathbf{NNZ}}>{\mathbf{N}}×\left({\mathbf{N}}-1\right)/2$ when ${\mathbf{DIRECT}}=\mathrm{.FALSE.}$, or ${\mathbf{NNZ}}<1$.
${\mathbf{IFAIL}}=3$
On entry, ${\mathbf{IROW}}\left(k\right)<1$ or ${\mathbf{IROW}}\left(k\right)>{\mathbf{N}}$ or ${\mathbf{ICOL}}\left(k\right)<1$ or ${\mathbf{ICOL}}\left(k\right)>{\mathbf{N}}$ or ${\mathbf{IROW}}\left(k\right)={\mathbf{ICOL}}\left(k\right)$ for some $k$ when ${\mathbf{DIRECT}}=\mathrm{.TRUE.}$.
${\mathbf{IFAIL}}=4$
On entry, ${\mathbf{IROW}}\left(k\right)<1$ or ${\mathbf{IROW}}\left(k\right)\ge {\mathbf{ICOL}}\left(k\right)$ or ${\mathbf{ICOL}}\left(k\right)>{\mathbf{N}}$ for some $k$ when ${\mathbf{DIRECT}}=\mathrm{.FALSE.}$.
${\mathbf{IFAIL}}=5$
${\mathbf{D}}\left(k\right)\le 0.0$ for some $k$.
${\mathbf{IFAIL}}=6$
On entry, ${\mathbf{IROW}}\left(k-1\right)>{\mathbf{IROW}}\left(k\right)$ or ${\mathbf{IROW}}\left(k-1\right)={\mathbf{IROW}}\left(k\right)$ and ${\mathbf{ICOL}}\left(k-1\right)>{\mathbf{ICOL}}\left(k\right)$ for some $k$.
${\mathbf{IFAIL}}=7$
On entry, ${\mathbf{IROW}}\left(k-1\right)={\mathbf{IROW}}\left(k\right)$ and ${\mathbf{ICOL}}\left(k-1\right)={\mathbf{ICOL}}\left(k\right)$ for some $k$.
${\mathbf{IFAIL}}=8$
No connected network exists between vertices NS and NE.

## 7  Accuracy

The results are exact, except for the obvious rounding errors in summing the distances in the length of the shortest path.

H03ADF is based upon Dijkstra's algorithm (see Dijkstra (1959)), which attempts to find a path ${n}_{s}\to {n}_{e}$ between two specified vertices ${n}_{s}$ and ${n}_{e}$ of shortest length $d\left({n}_{s},{n}_{e}\right)$.
The algorithm proceeds by assigning labels to each vertex, which may be temporary or permanent. A temporary label can be changed, whereas a permanent one cannot. For example, if vertex $p$ has a permanent label $\left(q,r\right)$, then $r$ is the distance $d\left({n}_{s},r\right)$ and $q$ is the previous vertex on a shortest length ${n}_{s}\to p$ path. If the label is temporary, then it has the same meaning but it refers only to the shortest ${n}_{s}\to p$ path found so far. A shorter one may be found later, in which case the label may become permanent.
The algorithm consists of the following steps.
1. Assign the permanent label $\left(-,0\right)$ to vertex ${n}_{s}$ and temporary labels $\left(-,\infty \right)$ to every other vertex. Set $k={n}_{s}$ and go to 2..
2. Consider each vertex $y$ adjacent to vertex $k$ with a temporary label in turn. Let the label at $k$ be $\left(p,q\right)$ and at $y\left(r,s\right)$. If $q+{d}_{ky}, then a new temporary label $\left(k,q+{d}_{ky}\right)$ is assigned to vertex $y$; otherwise no change is made in the label of $y$. When all vertices $y$ with temporary labels adjacent to $k$ have been considered, go to 3..
3. From the set of temporary labels, select the one with the smallest second component and declare that label to be permanent. The vertex it is attached to becomes the new vertex $k$. If $k={n}_{e}$ go to 4.. Otherwise go to 2. unless no new vertex can be found (e.g., when the set of temporary labels is ‘empty’ but $k\ne {n}_{e}$, in which case no connected network exists between vertices ${n}_{s}$ and ${n}_{e}$).
4. To find the shortest path, let $\left(y,z\right)$ denote the label of vertex ${n}_{e}$. The column label ($z$) gives $d\left({n}_{s},{n}_{e}\right)$ while the row label ($y$) then links back to the previous vertex on a shortest length ${n}_{s}\to {n}_{e}$ path. Go to vertex $y$. Suppose that the (permanent) label of vertex $y$ is $\left(w,x\right)$, then the next previous vertex is $w$ on a shortest length ${n}_{s}\to y$ path. This process continues until vertex ${n}_{s}$ is reached. Hence the shortest path is
 $ns→…→w→y→ne,$
which has length $d\left({n}_{s},{n}_{e}\right)$.

## 9  Example

This example finds the shortest path between vertices $1$ and $11$ for the undirected network
Figure 1