DEVELOPMENT OF ALGORITHMS TO SOLVE COMBINATORIAL

PROBLEMS

Broderick Crawford, Carlos Castro

Pontiﬁcia Universidad Cat

olica de Valpara

ıso, Valpara

ıso, Chile

Universidad T

ecnica Federico Santa Mar

ıa, Valpara

ıso, Chile

Eric Monfroy

LINA, Universit

e de Nantes, Nantes, France

Keywords:

Software Engineering, Agile Development, Knowledge Management, Creativity, Optimization Algorithms.

Abstract:

In this paper we present important human and social factors in relation with the development of algorithms to

solve combinatorial problems using different techniques. We ﬁx some concepts from knowledge management

and software engineering as applied developing optimization algorithms, solvers and metaheuristics.

1 INTRODUCTION

The development of algorithms to solve optimization

problems requires more knowledge than any single

person can possess. It requires the collaboration of

numerous individuals with complementary skills. The

necessary resources to solve problems are distributed

among the stakeholders and creative solutions emerge

out of collaborative work. Creative thinking is an area

that has been ignored in the development of Optimiza-

tion Algorithms. Nevertheless, its successful applica-

tion in the real world depends on a high degree of

creativity and innovation (Vidal, 2005). The develop-

ment of Optimization Algorithms and Metaheuristics

to solve Combinatorial Problems assumes the same

connotations it assumes in the ﬁeld of Software Engi-

neering. Then, the software development life cycle of

them might be quite diverse and different models from

other ﬁelds can be appropriate. This paper captures

our experience with valuable concepts from Knowl-

edge Management and Software Engineering applied

when we are developing algorithms.

In Software Engineering, the traditional ap-

proaches (often referred to as plan-driven, task-based

or Tayloristic), like the waterfall model and its vari-

ances, facilitate knowledge sharing primarily through

documentation. They also promote usage of role

based teams and detailed plans of the entire software

development life-cycle. It shifts the focus from in-

dividuals and their creative abilities to the processes

themselves. In contrary, agile methods emphasise and

value individuals and interactions over processes.

The agile principles and values have realized the

importance of collaboration and interaction in the

software development and, by other hand, creative

work commonly involves collaboration in some form

and it can be understood as an interaction between an

individual and a sociocultural context, the study of the

potential of techniques to foster creativity in software

engineering is a very interesting issue (Gu and Tong,

2004).

There are few studies reported on the importance

of creativity in software development. In manage-

ment and business, researchers have done much work

about creativity and obtained evidence that the em-

ployees who had appropriate creativity characteris-

tics, worked on complex, challenging jobs, and were

supervised in a supportive, noncontrolling fashion,

produced more creative work. Since human creativity

is thought as the source to resolve complex problem

or create innovative products, one possibility to im-

prove the software development process is to design

a process which can stimulate the creativity of devel-

opers.

We believe that in Optimization Algorithms de-

velopment projects, a better understanding of some

value and interdisciplinary concepts from Creative

Solving Problem (Vidal, 2005) and Knowledge Man-

agement (Nonaka and Takeuchi, 1995) offers impor-

tant insights about the use of Software Engineering

methodologies.

636

Crawford B., Castro C. and Monfroy E. (2007).

DEVELOPMENT OF ALGORITHMS TO SOLVE COMBINATORIAL PROBLEMS.

In Proceedings of the Ninth International Conference on Enterprise Information Systems - ISAS, pages 636-639

DOI: 10.5220/0002407006360639

 SciTePress

2 KNOWLEDGE MANAGEMENT

IN SOFTWARE ENGINEERING

The main argument to Knowledge Management in

software engineering is that it is a creative and knowl-

edge intensive activity. Software development is a

process where every person involved has to make a

large number of decisions and individual knowledge

has to be shared and leveraged at a project and or-

ganization level, and this is exactly what KM pro-

poses. People in such groups must collaborate, com-

municate, and coordinate their work, which makes

knowledge management a necessity. In software de-

velopment one can identify two types of knowledge:

Knowledge embedded in the products or artifacts,

since they are the result of highly creative activi-

ties and Meta-knowledge, that is knowledge about

the products and processes. Some of the sources

of knowledge (artifacts, objects, components, pat-

terns, templates and containers) are stored in elec-

tronic form.

However, the majority of knowledge is tacit, resid-

ing in the brains of the employees. A way to address

this problem can be to develop a knowledge sharing

culture, as well as technology support for knowledge

management. There are several reasons to believe

that knowledge management for software engineering

would be easier to implement than in other organiza-

tions: technology is not be intimidating to software

engineers and they believe the tools will help them do

a better job; all artifacts are already in electronic form

and can easily be distributed and shared; and the fact

that knowledge sharing between software engineers

already does occur to a large degree in many success-

ful software collaborative projects (Rus and Lindvall,

2002).

3 AGILE MANIFESTO

A new group of software development methodolo-

gies has appeared over the last few years. For a

while these were known as lightweight methodolo-

gies, but now the accepted term is Agile method-

ologies. The most common of them are: eX-

treme Programming, the Crystal Family, Agile Mod-

eling, Adaptive Software Development, Scrum, Fea-

ture Driven Development, Dynamic System Develop-

ment Method (Fowler, 2001). There exist many varia-

tions, but all of them share the common principles and

core values speciﬁed in the Agile Manifesto (Chau

and Maurer, 2004). Through this work they have

come to value individuals and interactions over pro-

cesses and tools. Working software over compre-

hensive documentation. Customer collaboration over

contract negotiation. Responding to change over fol-

lowing a plan. These new methods attempt a useful

compromise between no process and too much pro-

cess, providing just enough process to gain a reason-

able payoff. The result of all of this is that agile meth-

ods have some signiﬁcant differences with the former

engineering methods (Fowler, 2001): Agile methods

are adaptive rather than predictive and they are people

oriented rather than process oriented.

Most agile methodologies assume that change is

inevitable, these methodologies have the ability to ad-

dress variance and adaptability within the processes.

In (Highsmith and Cockburn, 2001) Highsmith and

Cockburn have ﬁxed the role of creativity in agile

teams assuming a world view that organizations are

complex adaptive systems. A complex adaptive sys-

tem is one in which decentralized, independent in-

dividuals interact in selforganizing ways, guided by

a set of simple, generative rules, to create innova-

tive, emergent results. Agile methods offer generative

rules, a minimum set of things you must do under all

situations to generate appropriate practices for special

situations. A team that follows generative rules de-

pends on individuals and their creativity to ﬁnd ways

to solve problems as they arise. Creativity, not volu-

minous written rules, is the only way to manage com-

plex software development problems and diverse sit-

uations.

4 SOFTWARE CREATIVITY

There are many deﬁnitions of creativity, we use some

ideas from (Franken, 2002): Creativity is deﬁned as

the tendency to generate or recognize ideas, alterna-

tives, or possibilities that may be useful in solving

problems, communicating with others, and entertain-

ing ourselves and others. There are three reasons why

people are motivated to be creative:

• need for novel, varied, and complex stimulation

• need to communicate ideas and values

• need to solve problems

In order to be creative, you need to be able to view

things in new ways or from a different perspective.

Among other things, you need to be able to generate

new possibilities or new alternatives. Tests of creativ-

ity measure not only the number of alternatives that

people can generate but the uniqueness of those alter-

natives. the ability to generate alternatives or to see

things uniquely does not occur by change; it is linked

to other, more fundamental qualities of thinking, such

DEVELOPMENT OF ALGORITHMS TO SOLVE COMBINATORIAL PROBLEMS

637

as ﬂexibility, tolerance of ambiguity or unpredictabil-

ity, and the enjoyment of things heretofore unknown.

Toward an understanding of creativity in organiza-

tions, the use of a creativity management framework

may be useful. Amabile (Amabile, 1996) had pro-

posed a theory for the development of creativity. In

her framework, creativity is hypothesized as a conﬂu-

ence of three kinds of resources:

• creativity-relevant skills (across domains)

• domain-relevant knowledge and skills (domain-

speciﬁc)

• task motivation

Domain-relevant resources include factual knowl-

edge, technical skills and special talents in the do-

main. Creativity-relevant resources include appropri-

ate cognitive style, personality trait, conducive work

style and knowledge of strategies for generating novel

ideas. In speciﬁc, the major features of the appropri-

ate cognitive style are the preference of breaking per-

ceptual set and cognitive sets, keeping response op-

tions open, suspending judgment, etc. Furthermore,

Amabile had proposed that intrinsic motivation was

conducive to creativity; whereas extrinsic motivation

was detrimental. Concerning the nurturing of intrin-

sic motivation, she and others highlighted the impor-

tance of promoting a playful attitude in the environ-

ment. Persons who are able to maintain playfulness,

may continue to focus on the interest and enjoyment

they derived from the task. They are more likely to

keep their intrinsic motivation, even under external

constraints.

The importance of creativity has been investi-

gated in all the phases of software development pro-

cess (Glass, 1995; Gu and Tong, 2004) and focused in

the requirements engineering too (Robertson, 2005;

Maiden and Robertson, 2005; Mich et al., 2005).

Nevertheless, the use of techniques to foster creativ-

ity in requirements engineering is still shortly investi-

gated. It is not surprising that the role of communica-

tion and interaction is central in many of the creativ-

ity techniques. The most popular creativity technique

used for requirements identiﬁcation is the classical

brainstorming and more recently, role-playing-based

scenarios, storyboard-illustrated scenarios, simulat-

ing and visualizing have been applied in an attempt

to bring more creativity to requirements elicitation.

These techniques try to address the problem of iden-

tifying the viewpoints of all the stakeholders (Mich

et al., 2005). However, in requirements engineering

the answers do not arrive by themselves, it is neces-

sary to ask, observe, discover, and increasingly create

requirements. If the goal is to build competitive and

imaginative products, we must make creativity part of

the requirements process. Indeed, the importance of

creative thinking is expected to increase over the next

decade (Maiden and Gizikis, 2001).

The industrial revolution replaced agriculture as

the major economic activity, and then information

technology replaced industrial production. Now, the

information technology will be replaced with a new

dominant economic activity focusing on creativity:

The Conceptual Age. According to (Pink, 2005) we

are moving from High Tech to High Touch and High

Concept. The skill of storytelling is now a mandatory

business skill. The workers in highest demand will

be those with great social skills and a strong drawing

portfolio. With the prevalence of search engines, facts

are abundant and free, what is in demand now is the

ability to put those facts in order and in context. The

shift of IT organizations toward the creative sector

and companies striving to design innovative products

that combine and use existing technologies in unan-

ticipated ways is beginning to justify this prediction.

In (Robertson, 2005; Robertson, 2002) very in-

teresting open questions are proposed: Is inventing

part of the requirements activity? It is if we want

to advance. So who does the inventing? We cant

rely on the customer to know what to invent. The

designer sees his task as deriving the optimal solu-

tion to the stated requirements. We cant rely on pro-

grammers because theyre too far removed from the

clients work to understand what needs to be invented.

Requirements analysts are ideally placed to innovate.

They understand the business problem, have updated

knowledge of the technology, will be blamed if the

new product doesnt please the customer, and know if

inventions are appropriate to the work being studied.

In short, requirements analysts are the people whose

skills and position allows, indeed encourages, creativ-

ity.

In (Boden, 1990) the author, a leading authority on

cognitive creativity, identiﬁes basic types of creative

processes: exploratory creativity explores a possible

solution space and discovers new ideas, combinato-

rial creativity combines two or more ideas that already

exist to create new ideas, and transformational cre-

ativity changes the solution space to make impossible

things possible. Then, most Requirements Engineer-

ing activities are exploratory, acquiring and discover-

ing requirements and knowledge about the problem

domain. And the Requirements Engineering practi-

tioners have explicitly focused on combinatorial and

transformational creativity.

ICEIS 2007 - International Conference on Enterprise Information Systems

638

5 CONCLUSIONS

This paper was focused on some aspects of Knowl-

edge Management and Creativity in the context of

Optimization Algorithms development. The develop-

ment of Optimization Algorithms is a ﬁeld well suited

for creative studies, since it is a creative activity where

the problems often can only be solved through an iter-

ative process faciliting Knowledge Management and

exploration of new ideas.

Agile methods emphasis on people, communities

of practice, communication, and collaboration in fa-

cilitating the practice of sharing tacit knowledge at a

team level. They also foster a team culture of knowl-

edge sharing, mutual trust and care. Agile develop-

ment is not deﬁned by a small set of practices and

techniques. Agile development deﬁnes a strategic ca-

pability, a capability to create and respond to change,

a capability to balance ﬂexibility and structure, a ca-

pability to draw creativity and innovation out of a de-

velopment team, and a capability to lead organiza-

tions through turbulence and uncertainty. They rough

out blueprints (models), but they concentrate on creat-

ing working software. They focus on individuals and

their skills and on the intense interaction of develop-

ment team members among themselves and with cus-

tomers and management.

The agile principles and values have recognized

the importance of collaboration and interaction in the

software development team. Because creative work

commonly involves collaboration the study of tech-

niques to foster creativity in software engineering is

very interesting. Agile process to be helpful to gen-

erate novel and useful product. On the contrary, the

discipline based work are perceived to be useless to

produce novel products. The difference between them

is that creative work can motivate the generation of

something new.

Software development is a creative and knowl-

edge intensive process that involves the integration of

a variety of business and technical knowledge, an un-

derstanding from a Knowledge Management perspec-

tive offers important insights for designing and imple-

menting Optimization Algorithms and Metaheuris-

tics.

REFERENCES

Amabile, T. M. (1996). Creativity in Context: Update to the

Social Psychology of Creativity. Westview Press.

Boden, M. (1990). The Creative Mind. Abacus.

Chau, T. and Maurer, F. (2004). Knowledge sharing in ag-

ile software teams. In Lenski, W., editor, Logic ver-

sus Approximation: Essays Dedicated to Michael M.

Richter on the Occasion of his 65th Birthday, volume

3075 of Lecture Notes in Artiﬁcial Intelligence, pages

173–183. Springer.

Fowler, M. (2001). The new methodology. Available at

http://www.martinfowler.com/articles/

newMethodology.html.

Franken, R. E. (2002). Human Motivation. Thomson Learn-

ing College.

Glass, R. L. (1995). Software creativity. Prentice-Hall, Inc.,

Upper Saddle River, NJ, USA.

Gu, M. and Tong, X. (2004). Towards hypotheses on cre-

ativity in software development. In Bomarius, F. and

Iida, H., editors, PROFES, volume 3009 of Lecture

Notes in Computer Science, pages 47–61. Springer.

Highsmith, J. and Cockburn, A. (2001). Agile software

development: the business of innovation. Computer,

34(9):120–127.

Maiden, N. and Gizikis, A. (2001). Where do requirements

come from? IEEE Softw., 18(5):10–12.

Maiden, N. and Robertson, S. (2005). Integrating creativ-

ity into requirements processes: Experiences with an

air trafﬁc management system. In 13th IEEE Interna-

tional Conference on Requirements Engineering (RE

2005), 29 August - 2 September 2005, Paris, France,

pages 105–116. IEEE Computer Society.

Mich, L., Anesi, C., and Berry, D. M. (2005). Applying a

pragmatics-based creativity-fostering technique to re-

quirements elicitation. Requir. Eng., 10(4):262–275.

Nonaka, I. and Takeuchi, H. (1995). The Knowledge Creat-

ing Company. Oxford University Press.

Pink, D. (2005). A Whole New Mind: Moving from the

Information Age to the Conceptual Age. Riverhead

Hardcover.

Robertson, J. (2002). Eureka! why analysts should invent

requirements. IEEE Softw., 19(4):20–22.

Robertson, J. (2005). Requirements analysts must also be

inventors. Software, IEEE, 22(1):48–50.

Rus, I. and Lindvall, M. (2002). Knowledge man-

agement in software engineering. IEEE Soft-

ware, 19(3):26–38. Available at http://fc-

md.umd.edu/mikli/RusLindvallKMSE.pdf.

Vidal, V. V. (2005). Creativity for operational researchers.

Investigacao Operational, 25(1):1–24.

DEVELOPMENT OF ALGORITHMS TO SOLVE COMBINATORIAL PROBLEMS

639