Read Anytime Anywhere Easy Ebook Downloads at ebookmeta.

com

Advances in Optimization and Linear Programming

1st Edition Ivan Stanimirović

https://ebookmeta.com/product/advances-in-optimization-and-
linear-programming-1st-edition-ivan-stanimirovic/

OR CLICK HERE

DOWLOAD EBOOK

Visit and Get More Ebook Downloads Instantly at https://ebookmeta.com

 Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

Linear and Convex Optimization A Mathematical Approach 1st

Edition Michael H. Veatch

https://ebookmeta.com/product/linear-and-convex-optimization-a-
mathematical-approach-1st-edition-michael-h-veatch/

ebookmeta.com

Finite Element Programming in Non linear Geomechanics and

Transient Flow 1st Edition Nobuo Morita

https://ebookmeta.com/product/finite-element-programming-in-non-
linear-geomechanics-and-transient-flow-1st-edition-nobuo-morita/

ebookmeta.com

Recent Advances in Computational Optimization Results of

the Workshop on Computational Optimization WCO 2021 Stefka
Fidanova
https://ebookmeta.com/product/recent-advances-in-computational-
optimization-results-of-the-workshop-on-computational-optimization-
wco-2021-stefka-fidanova/
ebookmeta.com

Foundation Mathematics GCSE for Edexcel Student Book

Answers 1st Edition Karen Morrison

https://ebookmeta.com/product/foundation-mathematics-gcse-for-edexcel-
student-book-answers-1st-edition-karen-morrison/

ebookmeta.com
Understanding China through Big Data: Applications of
Theory-oriented Quantitative Approaches Yunsong Chen

https://ebookmeta.com/product/understanding-china-through-big-data-
applications-of-theory-oriented-quantitative-approaches-yunsong-chen/

ebookmeta.com

Parables for the Virtual Movement Affect Sensation 1st

Edition Brian Massumi

https://ebookmeta.com/product/parables-for-the-virtual-movement-
affect-sensation-1st-edition-brian-massumi/

ebookmeta.com

Data Science and Analytics for SMEs: Consulting, Tools ,

Practical Use Cases 1st Edition Afolabi Ibukun Tolulope

https://ebookmeta.com/product/data-science-and-analytics-for-smes-
consulting-tools-practical-use-cases-1st-edition-afolabi-ibukun-
tolulope/
ebookmeta.com

Double Or Nothing An Officially Approved All New 007

Trilogy Book I Kim Sherwood

https://ebookmeta.com/product/double-or-nothing-an-officially-
approved-all-new-007-trilogy-book-i-kim-sherwood/

ebookmeta.com

A Short and Happy Guide to the First Amendment 1st Edition

Robert Power Mark Alexander

https://ebookmeta.com/product/a-short-and-happy-guide-to-the-first-
amendment-1st-edition-robert-power-mark-alexander/

ebookmeta.com
Saunders 2022-2023 Clinical Judgment and Test-Taking
Strategies 7th Edition Linda Anne Silvestri

https://ebookmeta.com/product/saunders-2022-2023-clinical-judgment-
and-test-taking-strategies-7th-edition-linda-anne-silvestri/

ebookmeta.com
Advances in Optimization and
Linear Programming
Advances in Optimization
and Linear Programming

Ivan Stanimirović
First edition published 2022
Apple Academic Press Inc. CRC Press
1265 Goldenrod Circle, NE, 6000 Broken Sound Parkway NW,
Palm Bay, FL 32905 USA Suite 300, Boca Raton, FL 33487-2742 USA
4164 Lakeshore Road, Burlington, 2 Park Square, Milton Park,
ON, L7L 1A4 Canada Abingdon, Oxon, OX14 4RN UK

© 2022 Apple Academic Press, Inc.

Apple Academic Press exclusively co-publishes with CRC Press, an imprint of Taylor & Francis Group, LLC
Reasonable efforts have been made to publish reliable data and information, but the authors, editors, and publisher
cannot assume responsibility for the validity of all materials or the consequences of their use. The authors, editors,
and publishers have attempted to trace the copyright holders of all material reproduced in this publication and
apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material
has not been acknowledged, please write and let us know so we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or
utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including
photocopying, microfilming, and recording, or in any information storage or retrieval system, without written
permission from the publishers.
For permission to photocopy or use material electronically from this work, access www.copyright.com or contact the
Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. For works that
are not available on CCC please contact [email protected]
Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used only for
identification and explanation without intent to infringe.

Library and Archives Canada Cataloguing in Publication

Title: Advances in optimization and linear programming / Ivan Stanimirović.
Names: Stanimirović, Ivan, author.
Description: First edition. | Includes bibliographical references and index.
Identifiers: Canadiana (print) 20210285362 | Canadiana (ebook) 20210285389 | ISBN 9781774637401
(hardcover) | ISBN 9781774637418 (softcover) | ISBN 9781003256052 (ebook)
Subjects: LCSH: Linear programming—Mathematical models.
Classification: LCC T57.76 .S73 2022 | DDC 519.7/2—dc23
Library of Congress Cataloging‑in‑Publication Data
Names: Stanimirović, Ivan, author.
Title: Advances in optimization and linear programming / Ivan Stanimirović.
Description: First edition. | Palm Bay, FL : Apple Academic Press, 2022. | Includes bibliographical references
and index. | Summary: “This new volume provides the information needed to understand the simplex method,
the revised simplex method, dual simplex method, and more for solving linear programming problems.
Following a logical order, the book first gives a mathematical model of the linear problem programming
and describes the usual assumptions under which the problem is solved. It gives a brief description of classic
algorithms for solving linear programming problems as well as some theoretical results. It goes on to explain
the definitions and solutions of linear programming problems, outlining the simplest geometric methods, and
showing how they can be implemented. Practical examples are included along the way. The book concludes
with a discussion of multi-criteria decision-making methods. This volume is a highly useful guide to linear
programming for professors and students in optimization and linear programming”-- Provided by publisher.
Identifiers: LCCN 2021037727 (print) | LCCN 2021037728 (ebook) | ISBN 9781774637401 (hardcover) |
ISBN 9781774637418 (paperback) | ISBN 9781003256052 (ebook)
Subjects: LCSH: Linear programming. | Simplexes (Mathematics) | Mathematical optimization.
Classification: LCC T57.76 .S83 2022 (print) | LCC T57.76 (ebook) | DDC 519.7/2--dc23
LC record available at https://lccn.loc.gov/2021037727
LC ebook record available at https://lccn.loc.gov/2021037728

ISBN: 978-1-77463-740-1 (hbk)

ISBN: 978-1-77463-741-8 (pbk)
ISBN: 978-1-00325-605-2 (ebk)
About the Author

Ivan Stanimirović, PhD, is currently working as an Associate Profes-

sor at the Department of Computer Science, Faculty of Sciences and
Mathematics at the University of Niš, Serbia. He was formerly with
the Faculty of Management at Megatrend University, Belgrade, as a
Lecturer. His work spans from multi-objective optimization methods
to applications of generalized matrix inverses in areas such as im-
age processing and restoration and computer graphics. His current
research interests include computing generalized matrix inverses and
their applications, applied multi-objective optimization and decision
making, as well as deep learning neural networks. Dr. Stanimirović
was the Chairman of a workshop held at the 13th Serbian Mathe-
matical Congress, Vrnjačka Banja, Serbia, in 2014.

v
Contents

About the Author v

Preface ix

1 Introduction 1
1.1 Multiobjective Optimization . . . . . . . . . . . . . 9
1.2 Symbolic Transformations in Multi-Sector
Optimization . . . . . . . . . . . . . . . . . . . . . . 11
1.3 Pareto Optimality Test . . . . . . . . . . . . . . . . 14
1.4 The Method of Weight Coefficients . . . . . . . . . 17
1.5 Mathematical Model . . . . . . . . . . . . . . . . . 33
1.6 Properties of a Set of Constraints . . . . . . . . . . 43
1.7 Geometrical Method . . . . . . . . . . . . . . . . . 54

2 Simplex Method 63
2.1 Properties of Simplex Methods . . . . . . . . . . . . 63
2.2 The Algebraic Essence of the Simplex Method . . . 70
2.3 The Term Tucker’s Tables and the Simplex Method
for Basic Permissible Canonical Forms . . . . . . . 76
2.4 Algorithm of Simplex Method . . . . . . . . . . . . 79
2.5 Determination of the Initial Basic Permissible
Solution . . . . . . . . . . . . . . . . . . . . . . . . . 101

vii
viii Contents

2.6 Two-Phase Simplex Methods . . . . . . . . . . . . . 103

2.6.1 A Two-Phase Simplex Method That Uses
Artificial Variables . . . . . . . . . . . . . . 104
2.6.2 Two-Phase Simplex Method Without
Artificial Variables . . . . . . . . . . . . . . 109
2.7 BigM Method . . . . . . . . . . . . . . . . . . . . . 116
2.8 Duality in Linear Programming . . . . . . . . . . . 125
2.9 Dual Simplex Method . . . . . . . . . . . . . . . . . 133
2.10 Elimination of Equations and Free Variables . . . . 139
2.11 Revised Simplex Method . . . . . . . . . . . . . . . 144
2.12 Cycling Concept and Anti-Cyclic Rules . . . . . . . 149
2.13 Complexity of Simplex Methods and Minty-Klee
Polyhedra . . . . . . . . . . . . . . . . . . . . . . . 156

3 Three Direct Methods in Linear Programming 163

3.1 Basic Terms . . . . . . . . . . . . . . . . . . . . . . 164
3.2 Minimum Angle Method . . . . . . . . . . . . . . . 166
3.3 Dependent Constraints and Application of Game
Theory . . . . . . . . . . . . . . . . . . . . . . . . . 171
3.4 Algorithms and Implementation Details . . . . . . . 176
3.5 Direct Heuristic Algorithm with General Inverses . 177

Bibliography 183

Index 191
Preface

In the first part of this book, we give a mathematical model of lin-

ear problem programming and describe the usual assumptions under
which the problem is solved. Below is a brief description of classic
algorithms for solving linear programming problems and some theo-
retical results. Proofs of the theorem can be found in standard mono-
graphs and linear textbooks programming and optimization from the
literature list. Then, we will study the definition and solution of lin-
ear programming problems. That is what it entails first mathematical
model and basic definitions and basic properties of a set of permis-
sible solutions. We will then outline the simplest geometric method
for solving the linear problem programming and show how it can be
implemented.
In the second chapter, we will study the simplex method for solv-
ing the linear programming problem in general form, and we will
detail its stages and the dual simplex method. We will also consider
another version called the simplex method and the revised simplex
method to resolve the problem of numerical stability of the classi-
cal simplex method. Finally, we will point to the problem of cycling
and two ways to overcome this problem. We will also show that it is
a simplex algorithm of exponential complexity, despite its excellent
properties on practical problems.

ix
x Advances in Optimization and Linear Programming

The third chapter presents the results from the work [43, 54],
and deals with modifications and refinements of particular stages of
simplex methods. In Ref. [54], this algorithm was used, since it does
not require the introduction of artificial variables. In this chapter,
two algorithms were introduced to obtain an initial basic permissible
solution in phase I of the two-phase simplex algorithm (described in
Refs. [40] and [55]). A new rule is described for the choice of basic and
non-basic variables to select the variable that enters the base and the
variable that leaves the base. At the end of this chapter, we will detail
the implementation of the simplex and revised simplex method.
In the final chapter, the post-optimal analysis of simplex methods
is studied, and multi-criteria decision-making methods are listed, and
final remarks are provided.
Chapter 1
Introduction

Mathematical programming problems occur in different disciplines.

For example, a stock market manager has to choose the investments
that will generate the highest possible profit at an opportunity; this
puts the risk of large losses at a predetermined level. The production
manager organizes production at the factory so that the quantity
of products and the quality are maximum, and the consumption of
materials and time are minimum, with limited resources (number of
workers, machine capacity, and opening hours). A scientist makes a
mathematical model of the physical process which best describes a
particular physical occurrence, and has a finite number of measure-
ment results available. Also, the model should not be too complicated.
Operations research is a branch of mathematics consistent in the
use of mathematical, statistical, and algorithms in order to carry out
a process of decision-making models. It often deals with the study of
complex real systems, in order to improve (or optimize) the operation
of the same. Operations research enables analysis of decision-making,
given the scarcity of resources to determine how they can maximize
or minimize them.
In mathematics and computer science, an algorithm is a step-
by-step procedure used to perform calculations. Algorithms are used
for computing, data processing, and automatic locking. More pre-

1
2 Advances in Optimization and Linear Programming

cisely, an algorithm is an efficient method expressed by a final list

of defined instructions. Start the pipe from the initial step and the
entrance (which can be empty). The instructions describe the calcu-
lation, which after execution leads to a finite number of successive
steps, giving the output in the last step. The complexity of an al-
gorithm can be described as the number of primitive operations or
basic steps that need to be performed. For certain input data, see the
analysis of algorithms in Ref. [11].
The symbolic expression is a concept that denotes the use of com-
puter tools for the transformation of mathematical symbols. It can
be used to compute explicit results without numerical errors. That
is why symbolic expression always expressed applies to conditioned
problems. These can be rational functions or polynomials of one or
more variables. There are several different computer algebra soft-
ware packages that support symbolic computing, such as MATHEMAT-
ICA, MAPLE, MATLAB.
Traditional programming languages are procedural languages.
The procedural program is written as a list of instructions, which
are executed step-by-step by reading Zotos. Programs in procedural
computer languages such as C can be used in calculations, but are
limited in cost in understanding more complex algorithms, because
they give little information about the steps. Many researchers use
the ability to develop “rapid-prototype” code to test the behavior of
the algorithm before investing effort in developing code for the algo-
rithm in a procedural language. The approach with MATHEMATICA has
great advantages in researching over procedural programming avail-
able in MAPLE and in procedural languages. MATHEMATICA allows several
program paradigms: object-oriented, procedural, symbolic, and func-
Introduction 3

tional programming. Our main goal of the research was to develop

algorithms suitable for implementation both in MATHEMATICA and in
procedural programming languages.
Although linear programming is very applicable in practice, many
problems in practice cannot be adequately linearized without while
the drastically does not lose on accuracy. In this case, nonlinear pro-
gramming methods are applied. In addition to nonlinearity, in many
problems, it is necessary to find the optimum of more than one objec-
tive function. In that case, we need to solve the problem of multi-
objective optimization. If all , the goal functions have an optimum
at the same point, the problem is trivial and comes down directly
to the problem of nonlinear or linear programming. In practice, this
situation is very rare. There are several methods for rescuing prob-
lems multiobjective optimization [36]. Common to all these methods
is that the initial problem in the appropriate way is to reduce the
problem of linear and nonlinear programming.
Higher sectoral criteria optimization can be seen as a continuation
of research in the classroom (single-criteria) optimization along with
some extensions. Formally, the basic extension is the introduction of
a vector criterion function, which leads to the vector maximum prob-
lem. Essentially, it is necessary to expand the concept of optimality.
Considering the problem of the vector maximum, the concept of op-
timality is replaced by the concept of noninferiority (Pareto optimal-
ity). The notion of a general (unique) optimization criterion can be
introduced, which includes criterion functions and decision-makers.
The solution of the multiobjective optimization problem obtained ac-
cording to such a criterion is optimal. In this case, the notion of the
optimal solution from the classical optimization can be retained in the
4 Advances in Optimization and Linear Programming

higher sectoral one. However, these difficulties will only arise when
trying to formalize such a unique criterion. Therefore, two phases or
stages are used in multiobjective optimization. In the first phase, a
set of “better” solutions is determined on the basis of a vector crite-
rion function. And in the second phase, based on the preference of
the decision-maker, the final decision is adopted, which can be called
optimal. The set of decisions presented to the decision-maker should
contain a small number of decisions, which are non-inferior to the
given criterion functions. The problem of multi-sector optimization
occurs most often in the planning of complex systems; for example,
regional development, development of water or electricity systems,
urban planning, and preservation of the natural environment [38].
The higher securitization problem occurs in economics as a problem
of determining the market equilibrium [38]. A similar problem arises
as a problem of equilibrium in game theory. In game theory, games
are considered in which decision theory appears as “group decision
making” or decision making with several decision-makers.
In this chapter, we will present the problem of multi-sector op-
timization, as well as the means for its solution. We will first give
a definition of the problem of multiobjective optimization as well as
necessary terms for later consideration. Theoretically, we will process
and implement several classical methods of multi-criteria optimiza-
tion. Each of the described methods will be illustrated by one or
more examples. Implementation considerations methods are original
and are taken from the works [57].
Introduction 5

The general formulation of multiobjective optimization (MOO)

has the general form:

max Q(x) = Q1 (x), . . . , Ql (x), x ∈ Rn

p.o. fi (x) ≤ 0, i = 1, . . . , m (1.0.1)
hi (x) = 0, i = 1, . . . , k.

where; Q1 (x), . . . , Ql (x), f1 (x), . . . , fm (x), g1 (x), . . . , gm (x) are the

real functions of n variables contained in the vector x = (x1 , . . . , xn ).
This task looks for a solution to x that maximizes all 1 functions
of the target. That is why the task of multiobjective optimization
(MOO) is also called the vector optimization task. Because of sim-
plicity only maximization problems are considered here. It is known
that the task of minimization simply translates into the task of max-
imization by multiplication criterion functions with −1. All further
presented definitions and methods are possible to adjust and solve
the minimization task.
We say that X ⊆ Rn is a set of admissible shadows if:

X = {x|fi (x) ≤ 0, i = 1, . . . , m; hi (x) = 0, i = 1, . . . , k}.

Each admissible solution x ∈ X corresponds to a set of criteria

values function, i.e., vector Q(x) = (Q1 (x), Q2 (x), . . . , Ql (x)). In that
way, the set of admissible solutions is mapped to the criterion set,
i.e., S = {Q(x)|x ∈ X}.
In the following text, the following terms will be used:
- Marginal solutions of the MOO task is determined by by op-
timizing each of the goal functions individually over the given one
permissible set, i.e., solving l one objective tasks:
6 Advances in Optimization and Linear Programming

max Qj (x), x ∈ Rn
p.o. fi (x) ≤ 0, i = 1, . . . , m
hi (x) = 0, i = 1, . . . , k.

∗ (j)∗ (j)∗ (j)∗

We will mark marginal solutions with x(j) = (x1 , x2 , . . . , xn ),
∗
where; x(j) is the optimal solution obtained by optimizing the j-th
objective function over a given allowable set X.
- The ideal values of the objective functions, denoted by the prices
with Q∗j are the values of objective functions for marginal solutions

∗
Qj∗ = Qj (x(j) ), æ = 1, . . . , l.

- The ideal values of the goal functions determine the ideal point
in the criterion space, i.e., ideal value of the vector function

Q∗ = (Q∗1 , Q∗2 , . . . , Q∗l ).

- If there is a solution x∗ that simultaneously maximizes all func-

tions of the target, i.e.:

x∗ = {x|Qj (x) = Q∗j , j = 1, . . . , l},

then such a word is called perfect hay solution.

In most cases, the marginal word shadow differs and the perfect
word shadow does not exist. When a perfect solution exists, then it
is not really a MOO problem.
Introduction 7

It is very important to keep in mind that in real problems, the

goals are over always in collision, which means that not everyone can
be fully reached. It is not possible to strictly define the optimum
nor for every two words formally determine which is better than the
other. For this reason, the process of obtaining a solution requests
the influence of the decision maker (hereinafter DM). This is the best
thing for someone who has a deeper insight into the problem, and
according to whose request, the solution is approached. Bearers deci-
sions can be made, and then the problem can be further complicated
because of their different goals, participation in decision-making and
the degree of responsibility they are willing to take on.
The fact that the tasks of the MOO, as a rule, do not have a
perfect meaning to reconsider the concept of optimality and the defi-
nition of the optimal solution. The concept of Pareto optimality plays
a key role in this. It is an expansion of the known concept of opti-
mality used in classical single-criterion optimization.
Pareto optimum is defined as follows:

Definition 1.0.1. The admissible solution x∗ represents Pareto op-

timum if there is no other admissible solution x such that:

Qj (x) ≥ Qj (x∗ ) ∀j = 1, . . . , l

where at least one of the inequalities turns into a strict inequality >.

In other words, x is Pareto optimum if it would improve the value

of any target function caused a deterioration in the value of another
target function. For Pareto optimum, there are the following syn-
onyms: efficient, dominant, and non-dominant solution.
8 Advances in Optimization and Linear Programming

In addition to the Pareto optimum, weak and strict (strong)

Pareto optimums are defined.

Definition 1.0.2. The admissible solution x∗ is weak Pareto opti-

mum if there is no one another admissible solution x such that

Qj (x) > Qj (x∗ ) ∀j = 1, . . . , l.

In other words, x∗ is a weak Pareto optimum if it cannot at the

same time improve all target functions.

Definition 1.0.3. The Pareto optimal solution x∗ is strict Pareto

optimum if there is a the number β > 0 such that for each index
j ∈ {1, . . . , l} and for each x which satisfies the condition:

Qj (x) > Qj (x∗ )

there is at least one i ∈ {1, . . . , l} \ {j} such that:

Qi (x) > Qi (x∗ )

and that it is valid

Qj (x) − Qj (x∗ )
≥ β.
Qi (x∗ ) − Qi (x)
The strict Pareto optimum is distinguished by the Pareto solu-
tion that does not change causes excessive relative changes in goal
functions.
The relation between the described optimums is such that each set
of stricter Pareto optimum is a subset of weaker optimums, i.e., every
Pareto is the optimum at the same time a weak Pareto optimum, and
every strict Pareto optimum is also a Pareto optimum.
Introduction 9

1.1 Multiobjective Optimization

In multi-criteria optimization, several opposing goal functions
should be reduced to a minimum at the same time respecting the
given restrictions:

Max.: Q(x) = {Q1 (x), . . . , Ql (x)}, x ∈ Rn

p.o.: fi (x) ≤ 0, i = 1, . . . , m (1.1.2)
hi (x) = 0, i = 1, . . . , k.

It is possible to construct an interval (often called a constraint) in

( ref M OO); we simply denote by X. Thus, the whole X is defined
by X = {x|fi (x) ≤ 0, i = 1, . . . , m; hi (x) = 0, i = 1, . . . , k}.
As a consequence, the notation xX will indicate that x satis-
fies the inequality and equality of boundaries in (1.0.1). With xj ∗
we denote the point that maximizes the j-th function of the target
depending on the constraint x ∈ X.
In general, there is no special point that maximizes all target
functions at once. For these reasons, a possible point is constructed
as optimal if there is no possible point with the same or goal function
being estimated. So that the true increase has the minimum value
of the target function. For the sake of completeness, we redefine the
definitions of non-inferior solution (Pareto-optimal solution) and ideal
(utopia) point from Refs. [12], [34], and [35].

Definition 1.1.1. Solution x∗ is said to be Pareto optimal solution

multiobjective optimization problems (1.0.1) if there is no other pos-
10 Advances in Optimization and Linear Programming

sible revision of x ∈ X so that Qj (x) ≥ Qj (x∗ ) for every j = 1, . . . , l,

and Qj (x) > Qj (x∗ ) for at least one index j.

Definition 1.1.2. A point x0 is the ideal point of the problem (1.0.1)

of the problem if and only if:

Qi (x0 ) = max{Qi (x)| x ∈ X}, i = 1, . . . , l.

x
The multicriteria problem is often solved by combining multiple
goals in one goal, which solution is Pareto optimal point for the orig-
inal problem.
The following optimization tools are available:
1. Compiled programming languages (Fortran 90, C, etc.) with
subroutine libraries (NAG, IMSL, etc.)
2. Interactive mathematical software: fast to define, solve, and
prototype small problems, less efficient for large ones: a. General tools
for numerical analysis (Matlab, IDL); b. Symbolic mathematical com-
puter systems (Mathematica, Maple, Macsyma); c. Modeling tools for
optimization (GAMS, AMPL).
One of the goals of our research was the implementation of the
main methods of multi-sector optimization in computer algebraic sys-
tem MATHEMATICA. The software package MATHEMATICA is one of the vari-
ous programming languages available today, and is applicable to sym-
bolic languages, and for numerical calculations Maeder, Wol. Several
functions for limiting the cost of numerical optimization are available
in the software package MATHEMATICA (see [33], [70]).
The Maximize and Minimize functions allow a specification of
the maximization and minimization goal function, along with a set
of constraints. In all cases, it is assumed that all variables are price
limits that do not have negative values.
Introduction 11

Minimize[f, {cons}, {x, y,...}] or Minimize[{f, cons}, {x,

y,...}], minimize f in the area specified by cons;
Maximize [f, {cons }, {x, y, ... }] or Maximize[{f,cons },
{x, y, ...}], find a maximum of f , in the area specified by cons.
Minimize and Maximize can be used to properly solve any polyno-
mial programming problem function in which the functions of the goal
f and the constraints cons include arbitrary polynomial functions of
the variables [70]. An important feature of Minimize and Maximize is
that they always find global minima and maxima [70].
The NMinimize and NMaximize functions implement several al-
gorithms for finding global optimum constraints. Expressions:
NMaximize[{f,cons}, vars, Method -> {method, mopts}],
NMinimize[{f,cons}, vars, Method -> {method, mopts}],
find the global maximum and minimum, respectively, for the objective
functions f by the constraint prices cons, using the method with
optimization methods method with the method options defined in
mopts.

1.2 Symbolic Transformations in Multi-Sector

Optimization

The main details of multi-sector optimization that are specific

to symbols and the expressions are described in the paper [57]. The
implementation was performed in the software package MATHEMATICA.
The method of weight coefficients, the main priority methods, and the
method of target programming are discussed. The symbolic conver-
12 Advances in Optimization and Linear Programming

sion of given goal functions and constraints into the corresponding

problem of one goal function is treated in particular. Transforma-
tions from multiobjective to one-criteria problem u procedural pro-
gramming languages are actually combinations of real values, and
involve procedures that depend on the function of the goal. In our im-
plementation, these transformations are performed in symbolic form
by taking combinations of target functions, which include undefined
symbols and unmarked variable prices.
We will suggest the following clear benefits that will result from
the implementation problems of multiobjective optimization in the
symbolic programming language MATHEMATICA, respect the traditional
implementation in procedural programming languages.

1. Possibility to use arbitrary target functions and limitations

(which are not defined by subroutines) during the execution
of the implementation function. The main aspects of these ad-
vantages are:
(i) The problem of secretory optimization (1.0.1) is represented
by a suitable in its form, whose elements can be used prices
as formal parameters in the optimization software. Inside the
dream form of the problem (1.0.1) is an edited triple

{Q1 (x), . . . , Ql (x)},

{f1 (x) ≤ 0, . . . , fm (x) ≤ 0, h1 (x) = 0, . . . , hk (x) = 0},
{x1 , . . . , xn } (1.2. 1)

The first element of the inner form, denoted by q, is a list

{Q1 (x), . . . , Ql (x) } whose elements indefinite expressions rep-
resenting goal functions. The second element in (1.2.1) is the
Introduction 13

constraint list fi (x) ≤ 0, i = 1, . . . , m, hi (x) = 0, i = 1, . . . , k.

We will label this argument as constr. The third element, la-
beled var, is a generic list of variables {x1 , . . . , xn }, determined
on the basis of x. In this sense, it is allowed that some argu-
ments in x can be defined in global environment MATHEMATICA
kernela.

(ii) If f is the objective function of the one-criteria optimization

problem obtained from (1.2. 1), we can calculate its maximum
using the standard function Maximize:
Maximize [f, constr, var].

The possibility of software to process arbitrary target functions

at arbitrary constraints enables the application of all optimiza-
tion models.

2. Possibility to use arrays of functions, whose elements can be

select and later apply to the given arguments. These structures
are not inherent in procedural programming languages.

Each expression contains in the list q = {Q1 (x), . . . , Ql (x)} can

be immediately applied to the given arguments. It is possible to cal-
culate the value of the function q[[[]]] in point x0 using the trans-
formation rules q[[i]] /. X - > X0 or q[[i]] /. Thread [Rule [x,
x0]]. MATHEMATICA searches for parts of the function q[[i]] that can
be replaced using the specified rule, and then performs a replacement.
Also, it is possible to define a function that takes another function
as an argument, for example: pf[f_, x_]: = f[x] + f[1-x]. Later,
expressions such as pf[q[[i]], x0] can be used. In the expression
f [x], the name of the function f is also expression, and can be treated
14 Advances in Optimization and Linear Programming

as any numeric expression [70]. In particular, any element from q can

be taken as an argument in the functions Minimize and Maximize.

3. An effective and natural symbolic transformation from a multi-

criteria model to a corresponding one-criterion model.

Many optimization methods are based on sub-algorithms that re-

quire the construction of goal functions of general form.

G (q(x), Φ(f (x), h(x), λ)) ,

where; λ is a possible set of larger variables, and Φ is an arbitrary

function.
Transformation of multiobjective to the one-criterion problem in
procedural languages is basically a combination of real values. We will
perform these transformations in symbolic form, using combinations
of unexpressed expressions and symbols.

1.3 Pareto Optimality Test

As a rule, it is impossible to find a complete infinite set of Pareto
optimal solutions to special problems from real life. For this reason,
the engineering securitization problem of the command seeks to de-
termine a subset of criterion-wise different Pareto optimal solutions
finally. Also, there are a number of methods for proving Pareto op-
timality. These methods can also be used to find the original Pareto
optimal solution of [?].
Introduction 15

An algorithm for determining the Pareto optimality was intro-

duced in the paper [57] solutions of multiobjective of the problem,
using direct proof in accordance with the Pareto definition of the
optimal point.

Algoritam 1.1 Pareto optimality test of fixed point x∗ .

Require: Optimization problem (1.0.1). Arbitrary fixed point x∗ .
1: Specify the set X = Reduce [constr /. List -> And, var] and
set Optimal = true.
2: For each index j = 1, . . . , l repeat Steps 2.1 and 2.2:
2.1: Generate the following conjunction constraint

P ar = X && u1 (x) && . . . && ul (x) (1.3.1)

where (
Qi (x) ≥ Qi (x∗ ), j 6= i,
ui (x) = (1.3.2)
Qi (x) > Qi (x∗ ), j = i.

2.2: If P ar = ∅, set Optimal := f alse and to Step 3.

3: return the value of the variable Optimal as a result.

The corresponding function in our implementation is:

IsPareto[q _List, constr _List, var _List, sol _List],
where the formal parameters are taken in the following sense:
q, constr, var: internal representation of the problem of sector
criteria optimization (1.0.1);
sol: The solution to the corresponding one-goal optimization
problem.
Calling this function is a form:
IsPareto[q, constr, var, First[Rest[Maximize[fun, constr,
var]]]];
16 Advances in Optimization and Linear Programming

where; fun, constr, and var are the representation of the corre-
sponding one-target optimization problem.
In this function we use the following version of the standard MATH-
EMATICA functions Reduce and FindInstance [70]:
Reduce [expr, var] finds all real values of variables contained
in the list var which satisfies a set of numbers containing logical
connections and invariant polynomial equations and inequalities.
FindInstance [expr, var] determines the values of the variables
from var in which the assertion is expr true. If no value is found
for var, the result is an empty list. The expr parameter can contain
equations, inequalities, a specification area, and quantifiers (see Ref.
[70]).

Algebra ‘InequalitySolve’
IsPareto[q_List, constr_List, var_List, res_List] : =
Module[{X = {}, l = Length[q]},
X = Reduce[constr/.{List->And}, var]; Ok = 1;
(* Korak 1. *)
For[j = 1, j < = l, j++, (* Korak 2. *)
Par = X;
For[i = 1, i < = l, i++, (* Korak 2.1. *)
If [j ! = i, Par = Par && q[[i]] > = (q[[i]]/. res),
Par = Par && q[[i]] > (q[[i]]/. res)
];
]; (* Par is of the form (4.1), (4.2) *)
If[FindInstance[Par, var] ! = {}, Ok = 0; Break[]];
(* Korak 2.2. *)
(*If a var instance is found, abort the loop*)
];
If[Ok = = 1,
Print[“Solution”, {q/.res, res}, “is Pareto optimal”],
Introduction 17

Print[“Solution”, {q/.res, res}, “is not Pareto optimal”];

];
Return[Ok]; (* Step 3. *)
]

1.4 The Method of Weight Coefficients

Weight coefficient method is the oldest method used for MOO.
According to this method, the weight coefficient wi is introduced for
all criterion functions Qi ( mathbf x), i = 1, ldots, l, so the problem
optimization reduces to the following scalar optimization:

l
X
max Q(x) = wi Qi (x) (1.4.1)
i=1
p.o. x ∈ X,

where; wi , i = 1, ldots, l meet the following conditions:

l
X
wi = 1, wi ≥ 0, i = 1, . . . , l.
i=1

The method of weight coefficients is often used by setting the val-

ues of these coefficients. However, this always causes certain difficul-
ties and objections to this procedure, because the subjective influence
on the final solution is entered through the given values of the weight
coefficients.
The main idea in the method of weight coefficients is to choose
weight coefficients wi that correspond to target functions Qi (x),
18 Advances in Optimization and Linear Programming

i = 1, . . . , l. Many authors have developed systematic approaches

in weight selection, the review of which can be found in Refs. [23],
[24], and [63]. One of the advantages of this method is that the vari-
ation of weights is consistent and continuous; it does not always have
to result in an accurate and complete representation of the Pareto
optimal set. This shortcoming has been discussed in Ref. [17].

Theorem 1.4.1. If all weight coefficients wi are positive, then the

solution of the problem (1.4.1) is Pareto optimal solution of the initial
MOO problem.

Proof. Let x∗ solve the problem (1.4.1), and let all weight coefficients
be strictly positive. Suppose that it is not Pareto optimal, i.e., there
exists x ∈ S so for i = 1, . . . , l it is valid Qi (x) ≥ Qi (x∗ ), where at
least one strict inequality holds (say for the index j). As wi > 0 for
every i, it is valid

l
X l
X
wi Qi (x) > wi Qi ( mathbf x∗ )
i=1 i=1

So we get a contradiction with the assumption that x∗ is a shadow

and problem (1.4.1). It follows that x∗ Pareto is optimal.

Theorem 1.4.2. If for every i ∈ {1, . . . , l} the condition wi ≥ 0 is

fulfilled, then the solution is problem (1.4.1) weak Pareto optimum of
the initial MOO problem.

Proof. Let x∗ solve the problem (1.4.1) and that the condition wi ≥ 0
is satisfied. Suppose that it is not weak Pareto optimal, i.e., yes there
is x ∈ S so that for i = 1, . . . , l it is valid Qi (x) > Qi (x∗ ). All
coefficients wi are non-negative and at least one is strictly greater
than zeros (due to li=1 wi = 1), so it is valid
P
Introduction 19

l
X l
X
wi Qi (x) > wi Qi (x∗ )
i=1 i=1

So we get a contradiction with the assumption that x∗ is a shadow

of the weight problem. Thus, x∗ is a weak Pareto optimum.

Theorem 1.4.3. If the solution of the problem (1.4.1) is unique, then

Pareto is also optimal.

Proof. Let x∗ be a unique solution to the problem (1.4.1). Suppose

that it is not a Pareto optimal solution to the MOO problem, i.e., that
there exists x ∈ S so that for i = 1, . . . , l it is valid Qi (x) ≥ Qi (x∗ ),
where at least one strict inequality (say for index j). Note that it
knows x 6= x∗ . Since wi ≥ 0 for every i, it is valid

l
X l
X
wi Qi (x) ≥ wi Qi (x∗ )
i=1 i=1

If a strict inequality were valid, then x∗ would not be a solu-

tion to the problem (1.4.1). So your equality. This means that there
are two different readings of x and x∗ problem (1.4.1), which is a
contradiction.

We will now show a strong assertion.

Theorem 1.4.4. If all wi > 0, i ∈ {1, . . . , l}, then the solution of

the problem (1.4.1) is a strict Pareto optimum of the MOO problem.

Proof. Let x∗ be the shadow of the weight problem. We have shown

that Pareto is optimal. Prove in winter that this solution is also a
strict Pareto optimum with a constant
20 Advances in Optimization and Linear Programming

wj
M = (k − 1) max .
i ,j wi
Suppose the opposite, that exists x ∈ S and index i such that
Qi (x) > Qi (x∗ ) where for each j for which Qj (x∗ ) > Qj (x) it is true
Qi (x∗ ) − Qi (x) < M (Qj (x) − Qj (x∗ )). With change:

(k − 1)wj
M=
wi
We get:

Qi (x∗ ) − Qi (x)
wi < wj (Qj (x) − Qj (x∗ )) > 0 .
k−1

So, for each j 6= i for which it is true that Qj (x∗ ) > Qj (x). For
index j =
6 i for which Qj (x∗ ) ≤ Qj (x) the above inequality certainly
holds. So, for each j 6= i it is true.

Qi (x∗ ) − Qi (x)
wi < wj (Qj (x) − Qj (x∗ )),
k−1
so by summing these inequalities for j = 1, . . . , i − 1, i + 1, . . . , l we
get:

l
X
wi (Qi (x∗ ) − Qi (x)) < wj (Qj (x) − Qj (x∗ ))
6
j=1,j=i

that is:

l
X l
X
∗
wj Qj (x ) < wj Qj (x) .
j=1 j=1
Introduction 21

We get that x∗ is not a solution to the weight problem, i.e., we are

coming to the contradiction. We conclude that x∗ is indeed a strict
Pareto optimum of the initial problem.

The weight coefficient should in some way represent the value of

the cross-function to which it is assigned. In order to achieve that, we
must first normalize the criterion functions, that is to them change
so that they have approximately equal values, while retaining all the
essential properties. For example, if it is a criterion function linear
Qj (x) = ni=1 ai xi , then normalized form of this function be:
P

Q (x)
P j .
i=1 nai

If the decision-maker himself defines the weight coefficients, then

this method belongs to the group of a priori methods. However, little
is usually known about how the odds should be chosen. Therefore, the
usual weight problem is solved for various vector values (w1 , . . . , wl )
and in this way, they get different readings between the sentences in
which DO chooses the one that suits him best. If this approach is
used, then this method becomes a posteriori method.
The main disadvantage of this method is the difficulty of deter-
mining the weight coefficients when we do not have enough informa-
tion about the problem. For this reason, we propose an algorithm
for the automatic generation of weight coefficients in order to obtain
Pareto optimal points. For positive weights and a convex problem,
the optimal solutions of a single-criterion problem are Pareto optimal,
i.e., minimizing the corresponding single-criteria problem is sufficient
for Pareto optimality.
22 Advances in Optimization and Linear Programming

There are numerous operations research problems investigated

and implemented at the moment. In [57], the implementation of poly-
nomial multi-objective optimization has been analyzed in MATHE-
MATICA. We developed a linear weighted sum method for multi-
objective optimization problem and presented it in [60]. Also, we
provided the so-called compendious lexicographic method for multi-
objective optimization tasks in [59].
Many applications of operations research exist in everyday life
and science and technology. For instance, an optimization algorithm
for LED-based Vis-NIR spectrally tunable light source is examined
in [32]. An application of the ELECTRE method to planetary gear
train optimization is detailed in [50], and in [51] a model of plan-
etary gear multicriteria optimization is further developed. We have
also investigated the computation of some efficient locations of the
Weber problem with barriers in [58]. Also, a heuristic algorithm for a
single resource-constrained project scheduling problem based on the
dynamic programming is investigated in [61].
Linear programming task is to determine the maximum (mini-
mum) linear function that depends on multiple variables provided
that these variables are nonnegative and satisfy linear restrictions in
the form of equations and/or inequalities. The linear objective func-
tion to be optimized is called the goal function or target function.
Linear programming is one of the most effective approaches to for-
mulating and solving complex decision-making problems and, as such,
is a fundamental discipline of operational research.
Linear programming does report in the field of solving practical
tasks such as are economic development planning both at the work
organization level so on the broader regional or broadest social plane.
Introduction 23

It is generally accepted that it is the first work to belong linear

programming published by L.V. Kantorovic in 1939 [26]. It defines the
transport task for the first time. The first work in this area included
the work of Colonel Vlastimir Ivanovic from 1940, entitled “Rules
for the calculation of the required number of means of transport”
[25]. In this paper, Colonel Ivanovic demonstrated how to calculate
the minimum number of vehicles for transporting a given quantity
of material using the “Principle of Economy.” Linear models are also
used in solving certain transport problems, which are known in theory
and practice as transportation problems, and in other planning areas.
The first algorithm for solving the transport problem was elab-
orated by F. F. Hitchock in [22] in 1941. On the formulation of the
transport problem and it’s also worked with Koopmans, which is the
results of his research published in 1947. The greatest contribution to
the development of linear of programming was given by G.B. Dantzig,
who formulated the general in 1947 linear programming problem and
set up a simplex method. In [16], the basics of simplex methods are
presented. The papers of John von Neumann from this period made
it possible in theory formulating a dual problem as well as finding a
connection between linear programming and game theory. Removal
method degeneration is suggested in [9]. Also significant are the pa-
pers of Gass and Saaty in 1955. Fr. parametric programming as well
as works by E. Beale and R. Gomory from 1958 on integer program-
ming.
– Linear Programming
The linear programming is a mathematical method or algorithm
by which an undetermined problem is solved, formulated by linear
equations, optimizing the objective function, also linear.
24 Advances in Optimization and Linear Programming

The linear programming is to optimize (minimize or maximize) a

linear function, which we call objective function, such that the vari-
ables of the function are subject to a number of constraints expressed
by a system of linear inequalities.
The variables are real numbers greater than or equal to zero. If it
is required that the resulting value of the variable is an integer, the
resolution procedure is called integer programming. This is the kind
of programming that will be used to solve the problem that concerns
us, because the solution will be “number of sheets” to cut a certain
way, and veneers are a whole variable type.
– Excel Solver
It serves for solving linear and nonlinear optimization. You can
also specify restrictions on integer decision variables. Solver can solve
problems with up to 200 decision variables, 100 explicit constraints,
and 400 simple (upper and lower bounds or integer restrictions on
variables decision). The literature is a link to an online manual in a
pdf file that explains how to use the excel solver function.
From the year 1949, a number of publications listed on the basis
of the theory of linear programming and its applications to various
branches of the economy. Deserve special mention for the decisive in-
fluence they had in the perfection and dissemination of these mathe-
matical techniques, the work and activities of the Cowles Commission
for Research in economics, the Rand Corporation, the Department of
Mathematics at Princeton University, and the Carnegie Institute of
Technology.
Moya (1998, p. 63) mentions that George B. Dantzig and another
group of associated researchers in 1947, accepted the request of mili-
tary authorities of the government of the United States, they set out
to investigate how it could apply mathematics and statistics to solve
Introduction 25

problems of planning and progression for purely military purposes. In

the same year, Dantzig and his colleagues raised for the first time the
basic mathematical structure of the linear programming problem.
Generally speaking, one can say that any phenomenon involving
a given nonnegative number of variables (i.e., variables whose value
is positive or zero) can be linked together by relations of inequality or
equality and reflect the limited or restrictions that the phenomenon
presents in order to optimize an objective can be formulated as a
mathematical programming model. If both the constraints and the
objective function can be stated by linear expressions, we are facing
a particular field of mathematical programming that denominates
“linear programming.”
In this case, the word “programming” does not refer to computer
programming; but it is used as a synonym for planning. Linear pro-
gramming deals with planning activities to obtain an optimal result,
that is, the result that best achieves the specified target (as the math-
ematical model) between alternative solutions.
Weber (1984, p. 718), the linear programming problem is about
maximizing or minimizing a linear function of several primary vari-
ables, called a goal function subject to a set of equalities or linear
inequalities called constraints, with the further proviso that none of
the variables can be negative. The latter can be seen when the prob-
lem requires, through the ingenious device of expressing the variable
of interest as the difference of two non-negative variables.
Briefly, it stated that linear programming is a mathematical
method of solving problems where the objective is to optimize (maxi-
mize or minimize) a result from selecting the values of a set of decision
variables, respecting restrictions pertaining to availability resources,
26 Advances in Optimization and Linear Programming

technical specifications, or other conditions that limit the freedom of

choice.
As if we have a particular interest in linear programming, we can
represent a production system using a model or matrix which include:

• Costs and revenues generated per unit of activity (objective

function).

• Contributions and input and output requirements for each ac-

tivity (coefficients input/output).

• Resource availability, technical specifications, and business to

respect (the right-side values of the constraints).

Specifically, linear programming is a mathematical method to analyze

and choose the best among many alternatives. In general terms, one
can think of programming as a means to determine the best way to
distribute a limited amount of resources in an attempt to achieve an
expressible target to maximize or minimize a certain amount.
The general model of a programming problem consists of two
linear important very parts: the objective function and constraints.
– The Linear Objective Function
The mathematical expression of the target is called the objec-
tive function, and the goal must be to maximize or minimize that
expression.
The aim can be the maximization of some input variables, which
may vary from gross or net income, depending on the model is struc-
tured. Linear programming can also be applied to problems of cost
minimization, and these programs are based on a different set of cri-
teria for optimization.
Introduction 27

C1, C2, ..., coefficients Cn are the cost coefficients (known) or in-
come, depending on the type of problem you’re seeing resole. More-
over, y1, y2..., yn are the decision variables (variables, or activity
levels) to be determined such that the target is reached within the
constraints faced by the problem.
Restrictions, expressed by linear inequalities, are composed of
technical coefficients (Aij), activities, or processes (yn), which are
also taken into account in the target function and also the cellular
levels or limitations (Bi). According to Beneke and Winterboer (1984,
p. 25), there are three basic types of restrictions: “higher than” (>=)
of “less than” (<=) or equal (=), and these can be classified due to
their nature:

• Resource Constraints or Entries: As such may include land,

capital, labor, and facilities.

• External Constraints: This kind includes concepts such as gov-

ernment area of land allocations, credit limits assigned to prod-
ucts or legal obligations.

• Subjective Restrictions: These are the restrictions imposed by

the operator. Limits may be difficult to define, but often are
real and significant in the planning process. Often restrictions
come from their own personal goals or business of the glider.
Among the limitations of this kind may include the following:

– Limitations on the level of credit that the glider is ready to

be utilized. It is often less than the amount lenders are dis-
placed to contribute. The typical motivation for such kind
of limitations is little explicit desire to avoid the hazards
of debt.
28 Advances in Optimization and Linear Programming

– Restrictions risk level activities involving aspects related

to highly variable income such as raising sheep or cattle.
– Minimum restrictions concerning the operator consider de-
sirable for not properly direct income such as maintaining
pure, dairy cows or crops to maintain the qualities of the
breed cows field.

As for the application that has linear programming, Moya (1998, 63)
indicates that some of the major problems that came to settle with
this tool are located in three areas: (1) production management, (2)
evaluation project, and (3) inversion and agricultural applications.
We indicate that this list does not exhaust in any way the options
that linear programming has proven to be an excellent tool to support
decision-making.
In all these situations, we can identify three common terms [5]:

1. There is a global size (goal) that wants to be optimized (profit,

the difference between model predictions and experimental
data).

2. In addition to the global goal, there are usually additional re-

quirements or constraints that must be met (limited risk, re-
sources, model complexity).

3. There are certain sizes so that if their values are selected “good,”
they are also satisfied objective and limitations. These quanti-
ties are called optimization variables or parameters.

So, in order to set the problem of mathematical programming, we

must:

1. Select one or more optimization variables;

Introduction 29

2. Select the objective function f ;

3. Form a constraint set.

After that, the class of problem to which the mathematicians obtained

ǐs identified the model belongs to, and the method for solving it is
selected.
There are several methods for solving linear programming prob-
lems [5, 8]. The geometric method is applicable to problems with a
number of variables n = 2 or when n−m = 2, where m is the number
of constraints. In principle, the linear programming problem can be
solved geometrically in the case of n = 3. The disadvantage of the ge-
ometric method is that it does not solve the general task of linear pro-
gramming, but only some special cases. Until the first general method
for solving linear programming, problems came American mathemati-
cian Danzing in 1947 [16]. This method is known as the simplex linear
programming method. He also formulated a general form of a linear
programming problem and gave an algorithm for solving it, known as
the simplex method. Dantzig’s work has circulated among experts
for many years and served as a basis for all subsequent considera-
tions of the problem linear programming. Although other methods
have been found in the meantime, this method is still in use today
through numerous modifications (see papers [6]). Both geometric and
simplex methods seek maximum (minimum) target functions at the
boundaries of the boundary region.
In the later period, alternative approaches to resolution emerge
as linear programming problems. We mention the works which are
used generalized inverses [45] as well as works by Conn [10] and Dax
[18] that do not use the simplex method. However, the practice has
shown that the simplex method is very effective, in 1972. V. Klee
Other documents randomly have
different content
“But what am I to tell my son when he comes back?” said the mother
anxiously.

“Oh! you must feign illness, and when he asks you what will save you, say
that nothing can save you but a suckling of the earth sow. When he goes to
get it, the sow will tear him in pieces.”

Well (but not particularly well!), when Víťazko returned from the chase,
bringing a buck for his mother, she groaned and complained: “Alas! my
dear son, your toil has been in vain. It is no use your bringing me this good
food; I cannot eat it, for I am deadly sick.”

“Alas! mother, you must not die. Only tell me what would cure you, and I
will bring it for you, even though it were from hell,” cried the good
Víťazko, for he loved his mother well.

“I can only be cured if I get the suckling of the earth sow.”

Víťazko did not wait; he took his beech-tree and set off in quest of the earth
sow. He wandered through the country, poor soul! for he did not know
where to go, till at last he came to a tower, and there he found Holy Sunday.

“Where are you going?” asked Holy Sunday.

“I am going to the earth sow to get one of her sucklings. My mother is ill,
but this will cure her.”

“My dear boy, it will be a hard task for you to get that piglet. However, I
will help you. Only you must follow my advice exactly.”

Víťazko promised that he would follow it exactly. So first she gave him a
long, sharp spit, and then she said:

“Go to the stable and take my horse. He will bring you to the place where
the earth sow lies buried in the earth. When you have come there you must
prick one of her pigs. The pig will squeak, and the sow, hearing it, will start
up and run round the earth in a moment. But she won’t see you or anybody
else, and so she will tell the pigs that if they squeak again she will tear them
to pieces. Then she will lie down to sleep, and then you must spit the pig
and run quickly away. The pig will be afraid to squeak, the sow won’t stir,
and my horse will carry you away.”

Víťazko promised to carry out her directions exactly. He took the spit,
mounted the magic horse, and it brought him swiftly to the place—far, very
far it was—where the earth sow lay buried in the earth. Víťazko pricked one
of the pigs, and it squeaked terribly. The sow started wildly up and ran
round the earth in one moment. But the magic horse did not move, so the
sow did not see him or anybody else, and she said angrily to the pigs:

“If one of you squeaks, I will tear you all to pieces at once.”

Having said this, she buried herself again.

At once Víťazko spitted the pig. It kept quiet and didn’t squeak at all, and
the magic horse began to fly, and it wasn’t long till they were home again.

“Well, Víťazko, how did it go?” asked Holy Sunday.

“Well, it went just as you said, and here is the pig.”

“Very well. Take it to your mother.”

Víťazko gave her back the spit; he led the magic horse back to its stall,
thanked Holy Sunday, and, hanging the pig from the beech-tree, made haste
to go home to his mother.

The mother and the griffin were feasting; they did not expect Víťazko, and
here he was. They ran away and discussed what they should do with him.

“When he has given you the pig, you must still pretend to be ill,” said the
griffin; “and when he asks you what will save you, tell him that only the
Water of Life and the Water of Death can cure you. If he goes in quest of
that, he is bound to perish.”
Víťazko came running to the castle full of joy. He gave the pig to his
mother, but she still went on groaning and complaining that she was going
to die, and that the pig would not cure her.

“Alas! mother, don’t die, but tell me what will cure you, so that I may bring
it for you at once,” said Víťazko anxiously.

“Ah! my dear son, I can only be cured by the Water of Life and the Water of
Death, and where would you get that?” sighed the mother.

Víťazko did not waste time thinking about it. He grasped his beech, and off
he went to Holy Sunday.

“Where are you going, Víťazko?” asked Holy Sunday.

“I am coming to you to ask where I could find the Water of Life and the
Water of Death, for my mother is still ill, and only those will cure her.”

“It will be a hard task for you to get them, but I will help you as well as I
can. Here are two jugs; mount my magic horse, and he will bring you to two
banks. Beneath those two banks spring forth the Water of Life and the
Water of Death. The right bank opens at noon, and from beneath it gushes
the Water of Life. The left bank opens at midnight, and beneath it is the
Water of Death. As soon as the bank opens, run up to it and fill your jug
with water, and so you must do in the other case too. When you have the
water, come back. Follow my instructions carefully.”

Saying this, she gave him two jugs. He took them and mounted the magic
horse, and in a moment they were gone like the wind. The two banks were
in a far distant land, and thither the magic horse brought Víťazko. At noon
he raised the right bank and the Water of Life gushed forth, then, crash! the
bank fell down again, and it was a wonder that it did not take Víťazko’s
heels off. Quickly Víťazko mounted the magic horse and made haste for the
left bank. There they waited till midnight. When the bank lifted, beneath it
was the Water of Death. He hurried to it and filled the jug, and, crash! down
fell the bank again; and it was a marvel it didn’t take Víťazko’s hand off.
Quickly he mounted the magic horse, the horse flew off, and soon they
were home again.

“Well, Víťazko, how have you fared?” asked Holy Sunday.

“Oh! everything went all right, Holy Sunday; and here is the water,” said
Víťazko, giving her the water.

Holy Sunday kept the water, and gave him two jugs full of spring water and
told him to take them to his mother. Víťazko thanked her and went home.

The mother and the griffin were carousing as before, for they did not expect
that he would ever return—and there he was just outside. They were terribly
frightened, and considered how they could get rid of him.

“You must pretend to be sick still, and tell him you won’t recover unless
you get the Pelican bird, and he will perish on the quest,” said the griffin.

Víťazko brought the water joyfully, but the mother was still groaning and
complaining; even that was no good, she was sure she was going to die.

“Ah! don’t die, sweet mother. Tell me what will cure you, and I shall be
glad to get it all for you,” said the good lad.

“There is no help for me unless I can see the Pelican bird. Where could you
get it for me?” groaned the mother.

Víťazko took his beech again, and it was no trouble to him to go to Holy
Sunday once more.

“Where are you going?” asked Holy Sunday.

“Well, I am coming to you to ask for advice. Mother is still sick; the water
did not cure her either, and she says she must see the Pelican bird. And
where is the Pelican bird?”

“My dear child, it would be very hard for you to get the Pelican bird. But I
will help you all I can. The Pelican bird is a gigantic bird. His neck is very
long, and, whenever he shakes his wings, he raises such a wind that the
trees begin to shake. Here is a gun; mount my magic horse, and he will
bring you to the place where the Pelican bird lives. But be careful. Point the
gun against the wind from whatever quarter it blows, and when the hammer
falls, ram the gun with the ramrod and come quickly back. You must not
look into the gun.”

Víťazko took the gun and mounted the magic horse, and the horse spread
his wings, and they were flying through the air a long way until they came
to a vast desert, where dwelt the Pelican bird. There the magic horse
stopped. Now Víťazko perceived that the wind was blowing strongly on his
left cheek, so he pointed the gun in that direction, and, clap! the hammer
fell. Víťazko rammed the gun quickly with the ramrod and flung it over his
shoulder, and the horse started flying, and very soon they were home again.

“Well, how did things go?”

“I don’t know whether they went well or ill, but I did what you
commanded,” answered Víťazko, handing down the gun to Holy Sunday.

“All right. You did quite right. Here he is!” she said. And then she took out
the Pelican bird. Then she gave Víťazko another gun to shoot an eagle with.
He went out into the forest, and returned before long with an eagle. She
gave him this eagle for his mother, in place of the Pelican bird.

The griffin and the mother were making merry again, hoping that Víťazko
would never come back, but he was already near. They were terrified, and
began to consider what new task they were to set him.

“You must pretend to be sick still, and tell him nothing can do you any good
but the golden apples from the garden of the Griffin. If he goes there the
Griffin will tear him in pieces, for he is enraged because Víťazko has killed
his brothers.”

Joyfully Víťazko gave the bird to his mother, but she still kept on groaning;
nothing was any good, only the golden apples from the garden of the Griffin
could save her.
“You shall have them,” said Víťazko, and without resting, he started again
and came to Holy Sunday.

“Where are you going, Víťazko?”

“Well,” he replied, “not even that did her any good. Mother is still sick, for
only the golden apples from the garden of the Griffin will cure her.”

“Well, you’ll have to fight, my boy,” said Holy Sunday; “but, even though
you were stronger than you are, it would be a bad look-out for you. Still, I
will help you all I can. Here is a ring for you; put it on your finger, and,
when you are in need, think of me, turn the ring round on your finger, and
you will have the strength of a hundred men. Now mount the magic horse;
he will take you there.”

Víťazko thanked her heartily, mounted the magic horse, and was carried by
him a far journey, till they came to a garden hedged about by a high
rampart. Had it not been for the magic horse Víťazko could never have got
into the garden, but the horse flew like a bird over the rampart. Víťazko
leapt down from the horse, and instantly began to look for a tree with
golden apples. A beautiful girl met him and asked him what he was looking
for. Víťazko said that he was looking for golden apples to cure his sick
mother, and begged her to tell him where to look for them.

“The apple-tree is under my charge, and I must not give the apples to
anybody, or the Griffin would tear me to pieces. I am a king’s daughter, and
the Griffin carried me off and brought me to this garden and put me in
charge of the apples. Go back, good youth, go back, for the Griffin is very
strong, and, if he sees you, he will kill you like a fly,” said the girl.

But Víťazko was not to be turned back, and he hastened on into the garden.
So the princess pulled off a priceless ring and handed it to Víťazko, saying:
“Take this ring, and when you think of me and turn this ring round on your
finger, you will have the strength of a hundred men, otherwise you could
not gain the victory over the Griffin.”
Víťazko took the ring and put it on his finger. He thanked her and went off
to the centre of the garden. In the middle of the garden stood an apple-tree
full of golden apples, and underneath it a horrible Griffin was lying.

“What do you want here, murderer of my brothers?” shouted the Griffin.

“I have come to get some apples from this tree,” answered Víťazko
undauntedly.

“You shall not have any of the apples unless you wrestle with me,”
exclaimed the Griffin angrily.

“I will if you like. Come on!” said Víťazko, and he turned the ring on his
right hand and thought of Holy Sunday. He set his legs wide apart and they
began to wrestle. In the first round the Griffin moved Víťazko a little, but
Víťazko drove him into the ground above his ankles. Just at this moment
they heard a swirl of wings above them, and a black raven shouted to them:

“Which am I to help, the Griffin or Víťazko?”

“Help me,” said the Griffin.

“And what will you give me?”

“I will give you gold and silver as much as you like.”

“Help me,” cried Víťazko, “and I will give you all those horses grazing on
yonder meadow.”

“I will help you, then,” said the raven. “But how am I to help you?”

“Cool me when I grow hot,” said Víťazko. He felt hot indeed, for the
Griffin was breathing out fire against him. So they went on wrestling. The
Griffin seized Víťazko and drove him into the ground up to his ankles.
Víťazko turned the ring, and again he thought of Holy Sunday. He put his
arms round the Griffin’s waist and drove him down into the ground above
his knees. The black raven dipped his wings in a spring, and then he
alighted on Víťazko’s head and sprinkled cool drops over Víťazko’s hot
cheeks, and thus he cooled him. Then Víťazko turned the other ring and
thought of the beautiful maiden, and they began wrestling again. So the
Griffin drove Víťazko into the ground up to his ankles, but Víťazko took
hold of him and drove him into the ground up to his shoulders, and quickly
he seized his sword, the gift of Holy Sunday, and cut the Griffin’s head off.

The princess came to him at once and plucked the golden apples for him.
She thanked him too for delivering her, and said that she liked him well and
she would marry him.

“I like you well too,” confessed Víťazko, “and, if I could, I would go with
you at once. But if you really love me, and if you will consent to wait a year
for me, I will come to you then.”

The princess pledged herself by shaking hands with him, and she said she
would wait a year for him. And so they said good-bye to each other.
Víťazko mounted his horse, cleared the rampart at a leap, killed the horses
on the meadow for the black raven, and hastened home.

“Well, how have you fared?” asked Holy Sunday.

“Very well, but if it hadn’t been for a ring which was given me by a
princess I should have fared very badly,” answered Víťazko, and he told her
everything. She told him to go home with the golden apples and to take the
magic horse with him too. Víťazko obeyed.

The griffin and the mother were carousing again. They were greatly startled
when Víťazko came riding home; they had never expected that he would
return alive even from the garden of the Griffin. The mother asked what she
should do; but the griffin had no more shifts; he made off to the tenth room
at once and hid himself there. When Víťazko had given the apples to his
mother, she pretended that the mere sight of them had cured her, and, rising
from the bed, she put the finest of food before Víťazko and then began to
caress him as she used to do sometimes when he was a tiny baby. Víťazko
was delighted to see his mother in good health again. The mother took a
strong cotton cord and said jestingly: “Lie down, dear son; I will wind this
cord round you as I used to wind it round your father, to see if you are as
strong as he was, and if you can break it.”

Víťazko smiled and laid himself down, and allowed his mother to wind the
cord round him. When she had finished, he stretched his limbs and snapt the
cord in pieces.

“You are strong,” she said. “But wait! I will wind this thin silk cord round
you to see if you can break it also.”

So she did. Víťazko tried to stretch his limbs, but the more he stretched, the
deeper the cord cut into him. So he was helpless, and had to lie like a baby
in its swaddling-clothes. Now the griffin hastened to cut his head off; he
hewed the body in pieces and hung the heart from the ceiling. The mother
packed the body in a cloth, and put the bundle on the back of the magic
horse, which was waiting in the courtyard, saying:

“You carried him alive, so you can carry him dead too, wherever you like.”

The horse did not wait, but flew off, and soon they reached home.

Holy Sunday had been expecting him, for she knew what would probably
happen to him. Without delay she rubbed the body with the Water of Death,
then she put it together and poured the Water of Life over it. Víťazko
yawned, and rose to his feet alive and well. “Well, I have had a long sleep,”
he said to himself.

“You would have been sleeping till doomsday if I hadn’t awakened you.
Well, how do you feel now?”

“Oh! I am all right! Only, it’s funny: it’s as though I had not got any heart.”

“That is true; you haven’t got a heart,” answered Holy Sunday.

“Where can it be, then?”

“Where else should it be, but in the castle, hanging from the crossbeam?”
said Holy Sunday, and she told him all that had happened to him.

But Víťazko could not be angry, neither could he weep, for he had no heart.
So he had to go and get it. Holy Sunday gave him a fiddle and sent him to
the castle. He was to play on the fiddle, and, as a reward, was to ask for the
heart, and, when he got it, he must return at once to Holy Sunday—those
were her orders.

Víťazko went to the castle, and when he saw that his mother was looking
out of the window, he began playing beautifully. The mother was delighted
with the music below, so she called the old fiddler (for Holy Sunday had put
that shape upon him) into the castle and asked him to play. He played, and
the mother danced with the griffin; they danced hard, and did not stop until
they were tired. Then the mother gave the fiddler meat and drink, and she
offered him gold, but he would not take it.

“What could I do with all that money? I am too old for it,” he answered.

“Well, what am I to give you, then? It is for you to ask,” said the mother.

“What are you to give me?” said he, looking round the room. “Oh! give me
that heart, hanging there from the crossbeam!”

“If you like that, we can give it to you,” said the griffin, and the mother
took it down and gave it to Víťazko. He thanked them for it, and hastened
from the castle to Holy Sunday.

“It is lucky that we have got it again,” said Holy Sunday; and she took the
heart in her hands, washed it first in the Water of Death and afterwards in
the Water of Life, and then she put it in the bill of the Pelican bird. The bird
stretched out his long neck and replaced the heart in Víťazko’s breast. At
once Víťazko felt it joyfully leaping. And for this service Holy Sunday gave
the Pelican bird his freedom again.

And now she said to Víťazko: “You must go once more to the castle and
deal out justice. Take the form of a pigeon and, when you think of me, you
will regain your own shape.”

No sooner had she said this than Víťazko was changed into a pigeon, and
away he flew to the castle. The mother and the griffin were caressing each
other when suddenly a pigeon alighted on the window-sill. As soon as the
mother saw the pigeon she sent the griffin to shoot him, but before the
griffin could get hold of his crossbow the pigeon flew down into the hall,
took human form, seized the sword and cut the griffin’s head off at a stroke.

“And what am I to do with thee, thou good-for-nothing mother?” he said,

turning to his mother, who in terror fell at his feet begging for mercy. “Do
not be afraid—I will not do you any harm. Let God judge between us.” He
took her hand and led her to the castle yard, unsheathed his sword, and said:
“Behold, mother! I will throw this sword into the air. If I am guilty, it will
strike me; if you are guilty, it is you it will strike. Let God decide.”

The sword whirled through the air, it darted past Víťazko’s head, and smote
straight into his mother’s heart.

Víťazko lamented over her and buried her. Then he returned to Holy
Sunday and thanked her well for all her kindness. He girded on the sword,
took his beech-tree in his hand, and went to his beautiful princess. He found
her with her royal father, who had tried to make her marry various kings
and princes, but she would marry none of them. She would wait a year, she
said. The year was not yet over when one day Víťazko arrived in the royal
palace to ask for the maiden’s hand.

“This is my betrothed,” exclaimed the princess joyfully, as soon as she saw

him, and she went straight up to him.

A splendid feast was made ready, the father gave his kingdom into their
hands, and that is the end of this story.
 BOOTS, CLOAK, AND RING

Once there was a blacksmith, and he had only one son, John by name. They
sent him to school, but fortune changed and his parents fell into poverty, so
they were forced to take their son home again. John had already passed
through the higher standard, but he could not support his parents. So one
day he said:

“Father and mother! What can I do at home? There is no business here, so I

can’t be a clerk, and I am too old now to learn a trade. So I will go out into
the world and find myself a job, and, whenever I can, I will send you some
money. And when I get a good job, you must sell your cottage and come
and live with me.”

His father and mother wept, because he wanted to leave them, but they
knew that he was right, for there was no chance for him if he stayed at
home. So they let him go. They gave him their blessing before he went out
into the world. John wept till his heart nearly broke at parting with his aged
parents.

He walked on till noon. At noon he sat down beneath a lime-tree beside a

well, and had his meal and a drink. Then, strengthened and refreshed, he
walked on till nightfall. The country was quite unknown to him, so he had
to spend the night in the forest. The next day he went on again till he came
into a wild mountain country. There he stopped and thought over what he
should do next. He stood awhile, and then he went on again. He reached a
pleasant valley, and there he found three brothers. They were quarrelling
and on the point of coming to blows. John asked them what the matter was.
The eldest answered:

“Our father has died, and he bequeathed to us these boots, this cloak, and
this hat. And each of us would like to own the boots.”

“Why?” asked John.

“Because they have the property that whoever puts them on can cover ten
miles in the moment he wishes it. The cloak has the property that its owner
can fly as far and as high as he likes. And the property of the hat is that it
makes its wearer invisible.”

John said: “You are brothers, and you ought not to quarrel. You must love
one another. So that you won’t quarrel any more, I will decide the matter for
you. Give me those things.”

They gave him the boots, the cloak, and the hat. He put the hat on, and they
couldn’t see him any more; he wrapped himself in the cloak, took the boots,
and flew away.

He flew some distance before he alighted upon a log and put the boots on.
As he sat on the log, it turned over, and he saw a big hole under it. He went
down the hole and came to some stairs, and went down them to the bottom
without any difficulty. There he found a big room without any human being
in it. The table was laid for one person. He thought: “I am hungry. Shall I
eat this meal?” Finally he decided to risk it; he took off his hat and began to
eat.

When he had finished, an old crone entered the room, and asked: “Did you
like your meal?”

“Oh, it was very nice indeed,” answered John; “and, by the way, could you
give me lodging for the night?”

“I will, if you can stand it; for at midnight twenty-four ghosts will come,
and they will try to make you play cards with them and dance with them.
But you must sit still and not so much as look at them.”

So the first night came. John was sitting eating his meal. When he had
finished, he remained at table. After eleven o’clock two dozen ghosts
entered the room and asked him to play cards with them. He refused, so
they began preparations for playing skittles, and again asked him to join
them, but he would not. Then a delightful music began to play, and they
asked him to dance with them. No, he wouldn’t; he did not so much as look
at them. They kept on dragging him about, tearing and biting him, till he
began to think it was all over with him. But just then it struck twelve, and
the ghosts vanished.

In the morning the old crone came back and waked him, for he was still
asleep on the ground. She asked him: “How did you sleep?”

“Very well,” said John.

“Did you, now?” answered the old woman. “Well, next night will be still
worse, if you can stand it. Two score of ghosts will come, and they’ll urge
you to play cards and skittles with them and to dance with them. But you
must sit quiet; don’t so much as look at them.”

He stayed there that day, and had a good time. Then the second night came.
After eleven o’clock twoscore ghosts rushed in. They urged him to play
cards and skittles with them and to dance with them. But John wouldn’t. He
sat still, without so much as looking at them. So they began to torture him
again, and dragged him about even worse than before. But when it struck
twelve they left him on the ground and disappeared.

In the morning the old crone came. She washed him with some lotion till he
recovered. She asked him: “How did you sleep?”

“Splendidly,” said he.

“Did you, now?” said she. “It was a bad lodging for you, but the third night
will be even worse, if you can stand it. Three score of ghosts will come, and
will urge you to play cards and skittles and to dance with them. But you
must sit still and not so much as look at them.”

All that day he had a good time again. The third night came, and after
eleven o’clock three score of ghosts rushed in. They gathered round him,
and urged, prayed, and besought him to play and dance with him. When he
refused, they seized him and began knocking him against the ground,
tearing and biting him, so that he lost his senses and did not see them go
away.
In the morning the old crone came and anointed him with a precious salve
till he recovered. The old woman said: “You wouldn’t have had such a bad
time if you had not stolen the boots, the cloak, and the hat. The ghosts
would simply have pressed you; they would have had no power over you.
As you followed my advice and did not play with them, you have delivered
an enchanted town and a beautiful princess. She will come to you at once.
Now you are rich, return the stolen goods.”

Then there came a girl in a white robe. It was the beautiful princess, and she
thanked him for delivering her and the whole town. He went to the window,
and outside he saw streets full of people and soldiers and a great bustle
going on. The princess said:

“My father is a king, and you will marry me and succeed him. But my
father dwells far from here, and we will go to him. Do you take this ring
here.”

So they went off. When the wedding was to take place, John wanted his
parents to be present, so he asked the princess: “May I go to see my
parents? I would like them to be at our wedding.”

The princess answered: “They live a great way from us, but you will be able
to get to them. The ring I gave you has the property that, when you turn it
on your finger and wish to go a hundred miles, you will cover that distance
in a moment. On your way you will come to a king who has a beautiful
daughter. But you must not think of her nor of me, for then you will lose the
ring, and you will not be able to go any farther.”

John started. He turned the ring, and in a moment he was a hundred miles
off, and found himself with a king who had several sons. They entertained
him splendidly. Then he came to another king who had an only daughter,
and she was very vulgar. The king insisted that John should marry her. John
thought: “What are you thinking of, my man? My bright one is so beautiful
that there is not her equal in the wide world, while your daughter is only a
vulgar creature.” At the moment he thought of his bride the ring slipped
from his finger and disappeared.
John left them then. He was very sad, and considered what he should do.
“My bride is far away now,” he thought. “I cannot find my way either to her
or to my parents.”

As he was walking along in this sad mood, he thought of his cloak, and it
came into his mind that, if he could reach the Sun’s abode the same day, he
could ask where his bride’s castle was. As soon as he thought of this he was
at the Sun’s house. The Sun was not in; only his housekeeper was at home.
He asked her for a lodging, and said that he would like to ask the Sun
whether he knew the castle where his bride dwelt. She gave him the
lodging. When the Sun returned home in the evening, John asked him
whether he had any knowledge of the castle in which his wife dwelt. The
Sun answered: “I don’t know. I never shone there. But go and ask the
Moon.”

The next day, as soon as he woke, he flew off on his cloak to the Moon’s
castle. When he got there, the Moon was not in, and John asked the
housekeeper for a night’s lodging. He said he would like to ask the Moon’s
advice.

The housekeeper said: “You must wait till the Moon comes home, but you
will be very cold, for my mistress is an extremely cold person.”

“I will crouch in a corner and wait till the mistress comes; in any case, my
cloak is warm enough.”

When the morning drew near, the Moon returned home, and John asked her
whether she knew where his bride’s castle was.

The Moon said: “I never shone there. But go to the Wind. He is a fellow
who penetrates everywhere, and so he is likely to know where that castle
is.”

So John went to the Wind’s house. The Wind was not in, but Melusine, his
wife, was alone at home. John asked her to let him stay there for the night.
She tried to dissuade him.
“It is impossible, good sir. My lord is used to blow terribly. It will be
exceedingly cold.”

He answered: “I will cover myself up and crouch somewhere. I can endure

cold, and, anyhow, my cloak is warm enough.” So he stayed there for the
night.

After midnight the Wind came home and asked: “Who is here with you,
wife? I smell a man.”

“Who should be here?” she said. “Your nose is still full of the human
smell.”

But the Wind persisted: “There is somebody here! Tell me!”

So she confessed. “Don’t be angry, dear husband! There is a man staying

here for the night, and he wants to ask you whether you will be kind enough
to take him to his bride’s castle.”

The Wind answered: “It is very far from here, and I must ask the Lord how
strongly I am to blow, if we are to get there. I was there yesterday; they are
going to celebrate a wedding there, and they have been drying some shirts
ready for it, and I have been helping them.”

The Wind went to ask the Lord; and when he came back, he said to John: “I
can blow strongly enough, but I don’t know whether you will be able to
keep step with me.”

John answered: “I have got good boots, and I am sure I can.”

So he wrapped himself in his cloak, covered his head with his hat, and put
his boots on, and he went ahead so quickly that the Wind could hardly keep
step with him. As they drew near to the castle, the Wind said: “Here it is,”
and disappeared in a whirl.

The other bridegroom had already arrived, and was at the wedding feast.
John passed through the castle, and came to the table at which they were
dining. Nobody could see him. He remained standing near the bride, and
whenever she lifted the food to her mouth, he ate it before it could reach her
mouth, so that the spoon reached her mouth empty.

After the banquet she said: “My plates were well filled, and yet it is as
though I had been eating nothing at all. Who is it that has eaten my food?
My glass was full too. I have not drunk, and yet it is empty. Who has drunk
my wine?”

Then she went to the kitchen, and John followed close at her heels. When
she was alone he revealed himself. He took his hat off, and she knew him.
She was greatly rejoiced at this, and ran to the room and said:

“Gentlemen, I would like to ask you a question. I had a golden key and I
lost it. So I had a silver key made for me, and, now that it is made, I have
found the golden key. Would you be so kind as to advise me which of them
I ought to keep?”

The bridegroom stepped forward and said: “Keep the golden key.”

Off she went. She dressed John in beautiful garments, and then presented
him to the guests, saying: “This is my golden key. He delivered me from
torment, and I was to marry him. He went to see his parents, but he could
not reach them. Now he has come back to me just as I was going to marry
another man, the silver key of my story, though I had given up all hope of
his return. Yet he has come back, and I shall keep him, the golden key, for
the silver key has himself decided so.”

The wedding was celebrated the next day, and John took charge of the old
king’s kingdom. Then they both went to visit his old parents, and brought
them back with them to the palace. On their way back they called on the
three brothers, and John gave them back the boots, the cloak, and the hat.
And if they haven’t died since, they are still alive enjoying their kingdom.
 SILLY JURA

Once there were two brothers. They were lazy fellows, and thieves into the
bargain. They were expected to give a feast. They said to one another: “We
haven’t got anything. Wherever shall we find food for the feast?”

So the first said: “I’ll go to our neighbour’s. He has some fine apples, and
I’ll pluck some of them.”

The second said: “I’ll go to the shepherd’s. He has some fine rams. I’ll steal
a ram from him.”

These two brothers hated the third, and so they abused him: “Silly Jura! You
won’t get anything yourself, but you’ll be ready enough to eat what we
get.”

So Jura said: “I’ll go to the burgomaster’s and get some nuts.”

In the evening they went their ways. When he had finished plucking the
nuts, Jura went into the charnel-house at the back of the church and began
to crack the nuts there. The watchmen heard the cracking in the charnel-
house, and they thought the place was haunted. As there was no priest in the
village (he lived in the next village), they went to the burgomaster and
asked him to go with them to the charnel-house, saying that the place was
haunted.

The burgomaster said: “I am so ill that I can’t stand on my feet; no doctor

can help me.”

But the watchmen insisted, and so the burgomaster told his servant to take
him on his back and carry him to the place. The servant carried him along,
and the watchmen called at the churchwarden’s to ask for some holy water.
When they came near to the charnel-house, Jura thought it was his brother
bringing the ram, so he called out: “Are you bringing him?”

The servant was frightened, and let the burgomaster fall and ran away. The
burgomaster was terrified too. He jumped up and ran after the servant. He
cleared a wooden fence with one leap in his flight, and it wasn’t long till he
reached home. His family wondered to see him cured so quickly without the
help of a doctor.

Next day the burgomaster proclaimed that he would give a pound to the
man who had stolen his nuts the day before, if he would only come to see
him. So Jura went to him, and the burgomaster said: “I ought to punish you
for stealing, but since you have cured my illness which nobody was able to
cure, I’ll give you the pound I promised, but you mustn’t steal any more.”
So Jura promised not to steal any more, and went home.

The brothers grew very fond of him now that he had money. They borrowed
the money from him and bought themselves new clothes, and said: “We’ll
go to see the world and to get wives for ourselves. As for you, Silly Jura,
you must stay at home; you’d never get a good wife for yourself.”

So off they went. But Jura went too. He went to the forest and he was
utterly dazzled. He had often heard that there was an enchanted castle in
that forest. When he came to the place where the ruins of the castle were,
night overtook him, and so he could see nothing except what looked like a
light in a cellar. So he went into the cellar to make his night’s lodging there.
There was nobody in the cellar but a cat. The cat greeted him: “Welcome,
dear Jura! How did you come here?”

Jura was frightened when he heard the cat speak, and was going to run
away. But the cat told him not to go; there was no need to be frightened. He
must come back, and no harm would be done to him. If he wanted to eat, he
could go into the storeroom and take what he wanted. She would take him
for her servant.

So he stayed there a year and had a good time. He never saw a cook, but he
always found meals ready prepared in the storeroom. He had nothing to do
but get firewood, and at the end of the year he was told to make a great pile
of it. Then the cat said: “You must light the pile to-day, and throw me into
the fire. You must not help me out, however I entreat you, but you must let
me be consumed.”

Jura answered: “I can’t do that. I have had a good time with you. Why
should I repay you in such an evil way?”

The cat said: “If you don’t do as I have said, you will be very unhappy. If
you do it, you will be happy.”

So Jura kindled the pile, and, when it was well alight, he picked up the cat
and threw her into the fire. She wanted to escape from the fire, but he
wouldn’t let her go. At last he was so weary that he was forced to lie down,
and soon he fell asleep. When he awoke, he opened his eyes, and behold!
there was no ruin; he heard delightful music and saw a beautiful palace with
crowds of servants. He was wondering at all this, when a splendidly dressed
lady came up to him and asked him if he did not know her.

Jura said: “How should I know your ladyship? I never saw you before in
my life.”

The lady said: “I am that cat. Witches had put me under enchantment in the
shape of a cat. Now we will go after your brothers who hated you so much
and see how they are getting on.”

She ordered her people to dress him in fine clothes, a fine carriage was
prepared, and they drove off. As they were approaching the village, the lady
said to her bridegroom: “Put your old clothes on.” Then she called an old,
ragged beggarwoman and sent him with her. She herself remained outside
of the village.

When the brothers saw Jura coming with the ragged beggarwoman, they
shouted: “He is bringing home an old ragged bride, and he’s in rags too.”
The other brothers were married too, and they were pretty badly off, so they
turned him out and wouldn’t have him at home.
So Jura went out of the village; he changed his clothes and drove back with
the lady to his brothers’ cottage. When the carriage stopped before the
cottage, the brothers said: “What a fine carriage! Who is that noble lord and
the beautiful lady who have come to our cottage?” They did not recognize
their brother.

So she said: “Look here. You were always hard on your brother, always
sneering at him, and now you are badly off enough, while he is getting on
splendidly. If you mend your ways, you will get on too.”

Afterwards she gave them some money and went away with Jura.
 SLEEPY JOHN

Once there was a lad named John, and he used to go to sleep always and
everywhere. One day he came to an inn where some farmers were feeding
their horses. So he crept into the cart, lay down on the straw, and went to
sleep. When the farmers had driven some distance, they noticed John asleep
in the cart. They thought: “What are we to do with him? We have a beer
cask here. We’ll put him in it and leave him in the forest.” So they shut him
in the cask, and off they drove.

John went on sleeping in the cask for a long time. Suddenly he woke up and
found himself in the cask, but he did not know how he had got into it,
neither did he know where he was. There was something running to and fro
near the cask, so he looked through the bunghole and saw a great number of
wolves gathered under the rocks. They had flocked round, attracted by the
human smell. One of the wolves pushed his tail through the hole, and
Sleepy John began to think that the hour of his death was approaching. But
he wound the wolf’s tail round his hand. The wolf was terrified, and,
dragging the cask after him, he ran after the rest of the wolves, who set off
in all directions. Their terror grew greater and greater as the cask bumped
after them. At last the cask struck against a rock and was smashed. John let
go the wolf, who took himself off as fast as he could.

Now John found himself in a wild mountain region. He began walking

about among the mountains and he met a hermit. The hermit said to him:
“You may stay here with me. I shall die in three days. Bury me then, and I
will pay you well for it.”

So John stayed with him, and, when the third day came, the hermit, who
was about to die, gave him a stick, saying: “In whatever direction you point
this stick, you will find yourself there.” Then he gave him a knapsack,
saying: “Anything you want you will find in this knapsack.” Then he gave
him a cap, saying: “As soon as you put this cap on, nobody will be able to
see you.”
Then the hermit died, and John buried him.

John gathered his things together, pointed the stick, and said: “Let me be
instantly in the town where the king lives.” He found himself there on the
instant, and he was told that the queen would every night wear out a dozen
pairs of shoes, yet nobody was able to follow her track. The lords were all
flocking to offer to follow the queen’s traces, and John went too. He went
into the palace and had himself announced to the king. When he came
before the king, he said that he would like to trace the queen. The king
asked him: “Who are you?”

He answered “Sleepy John.”

The king said: “And how are you going to trace her, when you are sleeping
all the time? If you fail to trace her you will lose your head.”

John answered that he would try to trace her all the same.

When the evening was come the queen went to bed in one room and John
went to bed in the next room, through which the queen had to pass. He did
not go to sleep, but when the queen was going by he pretended to be in a
deep slumber. So the queen lit a candle and scorched the soles of his feet to
make sure that he was asleep. But John didn’t stir, and so she was certain
that he was asleep. Then she took her twelve pairs of new shoes and off she
went.

John got up, put his cap on, and pointed with his stick and said: “Let me be
where the queen is.”

Now, when the queen came to a certain rock, the earth opened before her
and two dragons came to meet her. They took her on their backs and carried
her as far as the lead forest. Then John said: “Let me be where the queen
is,” and instantly he was in the lead forest. So he broke off a twig for a
proof and put it in his knapsack. But when he broke off the twig it gave out
a shrill sound as if a bell were ringing. The queen was frightened, but she
rode on again. John pointed with his stick and said: “Let me be where the
queen is,” and instantly he was in the tin forest. He broke off a twig again
and put it in his knapsack, and it rang again. The queen turned pale, but she
rode on again. John pointed with his stick again and said: “Let me be where
the queen is,” and instantly he was in the silver forest. He broke off a twig
again and put it into his knapsack. As he broke it, it gave out a ringing
sound and the queen fainted. The dragons hastened on again till they came
to a green meadow.

A crowd of devils came to meet them here, and they revived the queen.
Then they had a feast. Sleepy John was there too. The cook was not at home
that day, so John sat down in his place, and, as he had his cap on, nobody
could see him. They put aside a part of the food for the cook, but John ate it
all. They were all surprised to see all the food they put aside disappearing.
They couldn’t make out what was happening, but they didn’t care very
much. And when the banquet was at an end the devils began to dance with
the queen, and they kept on dancing until the queen had worn out all her
shoes. When her shoes were worn out, those two dragons took her on their
backs again and brought her to the place where the earth had opened before
her. John said: “Let me be where the queen is.” By this time she was
walking on the earth again, and he followed her. When they came near the
palace he went ahead of the queen and went to bed; and, as the queen was
going in, she saw him sleeping, and so she went to her own room and lay
down and slept.

In the morning the lords gathered together and the king asked whether any
of them had tracked the queen. But none of them could say “Yes.”

So he summoned Sleepy John before him. John said:

“Gracious Lord King, I did indeed track her, and I know that she used up
those twelve pairs of shoes upon the green meadows in Hell.”

The queen stood forth at once, and John took from his knapsack the leaden
twig and said: “The queen was carried by two dragons towards Hell, and
she came to the leaden forest; there I broke off this twig and the queen was
frightened.”

The king said: “That’s no good. You might have made the twig yourself.”
So John produced the tin twig from his knapsack and said: “After that the
queen drove through the tin forest, and there I broke off this twig. That time
the queen grew pale.”

The king said: “You might have made even this twig.”

So John produced the silver twig and said: “Afterwards the queen drove
through the silver forest, and when I broke off this twig she fainted, and so
she was until the devils brought her to life again.”

The queen, seeing that all was known, cried out: “Let the earth swallow
me!” and she was swallowed by the earth.

Sleepy John got the half of the kingdom, and, when the king died, the other
half too.