Inter-team Coordination in Large-scale Agile Software Development

143 Pages·2015·2.56 MB·English

by  Espen Andreassen

Inter-team Coordination in Large-scale
Agile Software Development
An Exploratory Case Study
Espen Andreassen
Master of Science in Computer Science
Submission date: June 2015
Supervisor: Torgeir Dingsøyr, IDI
Norwegian University of Science and Technology
Department of Computer and Information Science

Abstract
In later years agile development methodologies have seen a steady growth. Ag-
ile approaches were originally developed for small-scale contexts to cover the in-
creasing need for flexibility and the urge to be first-to-market with technology
in constant change. The benefits witnessed in this small-scale adoption has got
large organisations to open their eyes. Therefore, it has not been surprising to see
large-scale software development projects opt for the use of agile methodologies.
However, the research regarding agile development in a large-scale context is still
scarce.
Another aspect that has seen an increasing focus in the later years has been coor-
dination and coordination effectiveness, which are identified as important factors
in software development and team performance.
Thesetwoaspectsarecombinedandlookedfurtherintointhisresearchstudy. The
focus is on robust empirical studies performed on coordination in large-scale agile
software development projects, as well as an exploratory case study carried out
at a Norwegian large-scale agile software development project. The findings from
thiscasestudywasthencomparedtopreviouslypublishedresearchandtheoryon
coordination.
The main findings showed mainly similarities between existing studies and the
researchperformedinthismasterthesis. Themostprominentcoordinationmech-
anisms and aspects identified as having a positive effect oncoordination and team
performance were co-location, informal communication, presence of project man-
agementandowner,continuouschangeandimprovement,mutualtrust,andshared
mental models. However, some dissimilarities were also present. The most notice-
able of these were how informal communication arenas were existent to a large
degree in the case project despite some previously conducted research concluding
that such arenas should decrease when team and project size gets large.
Keywords: Large-scale; MTS; Multi-team systems; Coordination; Coordination
Effectiveness; Agile; SoftwareDevelopment; SoftwareEngineering; Scrum; Mutual
Trust; Shared Mental Models; Co-location; Informal Communication; Continuous
Change and Improvement; Presence from Management; On-site Customer

Sammendrag
I senere år har smidig utviklingmetodikk sett en stadig vekst. Smidige tilnær-
minger var opprinnelig utviklet for små-skala kontekster for å dekke det økende
behovetforfleksibilitetogtrangenetteråværeførstutepåmarkedetmedteknolo-
gier i konstant forandring. Fordelene identifisert i små-skala smidig utvikling har
åpnet øynene til større organisasjoner. Derfor har det ikke vært overraskende å
se at stor-skala smidige programvareutviklingsprosjekter velger å gå for smidig
utviklingmetodikk. Likevel viser det seg at forskning på smidige utviklingme-
todikker innenfor en stor-skala kontekst er mangelfull.
Et annet aspekt som har sett et økende fokus i de senere årene har vært ko-
ordinering og koordineringeffektivitet, som er identifisert som viktige faktorer i
programvareutvikling og oppnåelse av teamprestasjoner.
Disse to aspektene er kombinert og videre forsket på i dette forskningstudiet.
Fokuset vil være på solide empiriske studier tidligere utført på koordinering i
stor-skalasmidigprogramvareutviklingsprosjekter,itilleggtilenutforskendecase-
studie ved et norsk stor-skala smidig programvareutviklingsprosjekt. Funnene fra
denne case-studien vil bli sammenlignet med funn fra tidligere publisert arbeid og
teori på koordinering.
Hovedfunnene viste hovedsakelig likheter mellom eksisterende studier og arbei-
det utført i denne masteroppgaven. De mest fremtredende koordineringmekanis-
mene og aspektene identifisert til å ha en positiv effekt på koordinering og team-
prestasjonnivå var samlokalisering, uformell kommunikasjon, tilstedeværelse fra
prosjektledelse og prosjekteier, kontinuerlig endring og forbedring, gjensidig tillit,
ogfellesmentalemodeller. Detvarderimotnoenulikheteriforholdtileksisterende
studiertilstede. Denmestsynligeavdissevaratuformellekommunikasjonarenaer
i stor grad var eksisterende i case-prosjektet, selv om noe tidligere litteratur og
forskning hadde konkludert med at slike arenaer skulle minke i bruk når team- og
prosjektstørrelsen ble stor.
Nøkkelord: Stor-skala; MTS; Multi-team-systemer; Koordinering; Koordiner-
ingeffektivitet; Smidig; Programvareutvikling; Systemutvikling; Scrum; Gjensidig
Tillit; Felles Mentale Modeller; Samlokalisering; Uformell Kommunikasjon; Kon-
tinuerligEndringogForbedring; TilstedeværelsefraLedelse; SamlokalisertKunde

Preface
I am now fulfilling my last year on a master degree in computer science where
I specialise in software, or more specifically, software systems. I was introduced
to agile development methodologies through different subjects at the “Norwegian
University of Science and Technology”, NTNU, and also got hands-on experience
working with Scrum in a subject called “TDT4290 - Customer Driven Project”.
Thissubjectinparticularsparkedmyinterestinagiledevelopmentmethodologies
and the new ways of handling work and project organisation. After a summer
internship with EY (formerly known as Ernst&Young) I got more intrigued with
how communication and coordination was handled in real life business and IT
projects. Therefore, my previous experiences led to a motivation in exploring the
combination of agile development and coordination.
The work performed in this master thesis is carried out to give an insight in the
fieldofcoordinationinlarge-scaleagilesoftwaredevelopmentprojects. Ihopethat
the research performed will be beneficial for other research as the topic at hand is
still scarcely studied.
I would like to use this opportunity to thank Torgeir Dingsøyr for his support, as-
sistanceandknowledgethroughouttheresearchprojectasthesupervisor. Iwould
also like to thank NTNU for giving me the opportunity to experiment with ideas
within the boundaries of the research project, and letting me acquire interesting
knowledge for the future.
Trondheim, June 19, 2015
Espen Andreassen

Contents
I Introduction and Theory
1 Introduction 1
1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Problem Description and Background . . . . . . . . . . . . . . . . 2
1.3 Scope, Limitations and Acknowledgement . . . . . . . . . . . . . . 3
1.4 Target Audience . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.5 Report Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2 Theory 7
2.1 Software Development Methodologies. . . . . . . . . . . . . . . . . 8
2.2 Coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.3 Large-scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.4 Multiteam Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.5 Efficiency, Effectiveness, Productivity and Performance in Coordi-
nation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.6 Shared Mental Models . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.7 Mutual Trust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
II Research Methodology 43
3 Method 45
3.1 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.2 Research Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
III Results and Evaluation 55
4 Results 57
4.1 Clarification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.2 Overview of the Omega-project . . . . . . . . . . . . . . . . . . . . 58
4.3 Coordination Arenas and Important Aspects . . . . . . . . . . . . 65
5 Discussion 93
5.1 Research Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
5.2 Evaluation of the Study . . . . . . . . . . . . . . . . . . . . . . . . 102
IV Conclusion and the Road Forward 105
6 Conclusion 107
6.1 Research Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
7 Future Work 112
7.1 Suggestions for Future Research Focus . . . . . . . . . . . . . . . . 113
References i
V Appendices
A Interview Guide I
B Non Disclosure Agreement (NDA) IV