FTSProc: a Process to Alleviate the Challenges of Projects that Use the Follow-the-Sun Strategy

285 views

Published on

Paper presentation on ICGSE 2012

Published in: Technology
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
285
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
0
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

FTSProc: a Process to Alleviate the Challenges of Projects that Use the Follow-the-Sun Strategy

  1. 1. Pontifícia Universidade Católica do Rio Grande do Sul Faculdade de Informática Programa de Pós-Graduação em Ciência da Computação FTSProc: a Process to Alleviate the Challenges of Projects that Use the Follow-the-Sun StrategyEstevão Ricardo Hess Jorge Luis Nicolas AudyComputer Science School Computer Science SchoolPUCRS PUCRSPorto Alegre, Brazil Porto Alegre, Brazilestevaorh@gmail.com audy@pucrs.br
  2. 2. Agenda• Introduction• Related Works• FTSProc• The Experimental Process• FTSProc: Processo Proposto• Main Results• Conclusions 2
  3. 3. Introduction• Globalization process;• During global projects, the time zone difference is difficult to manage. However, this difference can also be used as an advantage;• In this sense, emerge the Follow-the-Sun (FTS) concept;• The FTS approach utilizes distributed team members spread across time zones to achieve a single project outcome. The main objective of FTS is to reduce the time-to-market in GSD environments; 3
  4. 4. Introduction• This paper shows a proposal process to be used during the development phase of the SDLC to alleviate the challenges presented in the FTS strategy;• This process focuses in the handoffs• The Process was named FTSProc and it was created based in the Composite Personae and the process called 24hr Design and Development;• It also uses the Test-driven development (TDD) technique;• Using the created process, this paper also presents an experiment used to validate the proposed process;• The findings during the experimental process brought evidences that a project that uses the FTSProc is more efficient then a project that does not use it; 4
  5. 5. Related Works• Fadel et al. – 2000 – 24hr Design and Development – To reduce the needed time to develop a software project – Experiment – distributed team in different time zones – Developed a work transfer process – Each development center uses 30 minutes to prepare the hand-off information – overlap – The team stating its shift do a planning meeting to define the job to be developed during the day • Brainstorming – This cycles finishes when all features have been developed• Denny et al. – 2008 – Composite Personae – Transitions based in the stand-up meetings – At the end of the day • All developers should add their results in the code repository • Fulfill an automated form – i. What have you done since last meeting? – ii. What are you planning to do until next meeting? – iii. Is there any impediments or blocker? – Next site continues the work based in the information providede by the previous team 5
  6. 6. FTSProc• This process aims to mitigate the challenges of coordination, synchronization and communication during the handoffs in the development phase of SDLC. In this sense, the main objectives of this process are: 1. When a team starts a working day (shift), it should simply have the perception of the work that must be developed and the work already done by previous development center. 2. Avoid the needed for synchronous communication between distributed teams. 3. Ensure that the handoff from a development center to another occurs without problems, and that the work can be continued from the point where the previous development center left off. 6
  7. 7. FTSProc 3. Coding 4. Check-in 2. Brainstorming Shift 5. Hand-off Form 1. Shift StartRequirements Design Feature implemented 7
  8. 8. FTSProc – Requirements• Requirements are defined as 3. Codingspecifically as possible[18], preferably using theconcept of User Stories 4. Check-in[19], having have well-definedacceptance criteria 2. Brainstorming Shift 5. Hand-off Form 1. Shift StartRequirements Design Feature implemented 8
  9. 9. FTSProc - Design• Artifacts showing how the 3. Codingrequirements will beimplemented 4. Check-in• TDD usage based in theacceptance criteria 2. Brainstorming Shift 5. Hand-off Form 1. Shift StartRequirements Design Feature implemented 9
  10. 10. FTSProc• Requirements clearly defined 3. Coding• Unit tests created for theTDD technique usage 4. Check-in• Development phase starts atthis point 2. Brainstorming Shift 5. Hand-off Form 1. Shift StartRequirements Design Feature implemented 10
  11. 11. FTSProc – Starts a shift• Starts a work day (shift) 3. Coding• Download the latest sourcecode from repository 4. Check-in• Generates a report with thecovered tests 2. Brainstorming• Generates a report with all Shiftinformation provided by theteam that worked on theprevious shift (Hand-off from) 5. Hand-off Form 1. Shift StartRequirements Design Feature implemented 11
  12. 12. FTSProc – Brainstorming• Daily planning 3. Coding• Task assignments• Assignments should consider 4. Check-inthe report with informationregarding the previous shift, as 2. Brainstormingwell as the result from the unittests report provided by the Shiftprevious site 5. Hand-off Form 1. Shift StartRequirements Design Feature implemented 12
  13. 13. FTSProc – Coding• This step represents therequirements implementation 3. Codingphase, following theassignments agreed during thebrainstorm. 4. Check-in• At this stage, the team 2. Brainstormingfocuses on the featuredevelopment. Shift 5. Hand-off Form 1. Shift StartRequirements Design Feature implemented 13
  14. 14. FTSProc – Check-In• Each team member mustperform the check-in of the 3. Codingwork done during theday, providing all necessaryinformation to the next team 4. Check-into continue the work where itwas left off. 2. Brainstorming Shift 5. Hand-off Form 1. Shift StartRequirements Design Feature implemented 14
  15. 15. FTSProc – Hand-Off Form• Each team member musttake time to fill out the hand- 3. Codingoff form• This form is based in the 4. Check-instund-up meetings 2. Brainstorming• This state represents the endof a shift Shift 5. Hand-off Form 1. Shift StartRequirements Design Feature implemented 15
  16. 16. FTSProc – Hand-Off Form• Each team member musttake time to fill out the hand- 3. Codingoff form• This form is based in the 4. Check-instund-up meetings 2. Brainstorming• This state represents the endof a shift Shifti. What has been done during the last period? 5. Hand-offii. How the work should be continued? Form 1. Shift Startiii. Is there any obstacle blocking the team?Requirements Designiv. What unit tests have been covered during this shift? Feature implemented 16
  17. 17. FTSProc – Feature completed• This process is interative and 3. Codingthese five steps will be repitedeach day (shift), for eachdistributed team 4. Check-in 2. Brainstorming Shift 5. Hand-off Form 1. Shift StartRequirements Design Feature implemented 17
  18. 18. Experimental process• Main objective: – To compare the efficiency of a project that utilizes the proposed process (FTSProc) to a project that does not utilizes (Ad hoc) – Quantity of requirements correctly implemented• Hypotheses – Null Hypothesis (H0): • H0: reqImplFTSProc = reqImplAdHoc – Alternative Hypothesis (H1): • H1: reqImplFTSProc > reqImplAdHoc – Alternative Hypothesis (H2): • H2: reqImplFTSProc < reqImplAdHoc 18
  19. 19. Experimental process• To evaluate the FTSProc: – Simple software development project – Mathematical System• Computer Science Master Students from PUCRS University• Professionals from the PUCRS’ technological park (TECNOPUC)• Sills requirements – Java – Test-driven Development (TDD) – JUnit – FTS – GSD 19
  20. 20. Experimental process• Participants distributions• 8 participants• 2 grupos • FTSProc X Ad hoc• 2 sites• Simulated time zones  One Day = 20 minutes• Lab  Keep control 20
  21. 21. Experimental process – Results Number of requirements correctly and partially implemented in each project: Implemented requirements FTSProc Ad hoc Correctly 12 4 Partially 0 8 Not implemented 0 0 Total 12 12 21
  22. 22. Experimental process – Results Time for each Shift FTSProc Ad hoc Hora Hora Hora Hora Atividade Tempo Total Atividade Tempo Total Inicial Final Inicial Final Análise dos Análise dos 18:05 18:10 5 Minutos 17:52 17:57 5 Minutos requisitos requisitos Shift 1 18:12 18:32 20 Minutos Shift 1 18:02 18:22 20 Minutos Shift 2 18:40 19:00 20 Minutos Shift 2 18:22 18:42 20 Minutos Shift 3 19:06 19:26 20 Minutos Shift 3 18:45 19:05 20 Minutos Shift 4 19:27 19:34 7 Minutos Shift 4 19:06 19:26 20 Minutos 22
  23. 23. Experimental process – Results The work handoff from one center to another occurred in an appropriate way? FTSProc Ad hoc Yes 4 3 No 0 1 Total 4 4 At the beginning of each shift, you could see directly how the work should be continued? FTSProc Ad hoc Yes 4 0 No 0 4 Total 4 4 Do you believe that the work handoff from a development center to another led to a significant work overhead? FTSProc Ad hoc Yes 0 1 No 4 3 Total 4 4 23
  24. 24. Results Discussion• After presenting the results and the evidence for the confirmation of the H1 hypothesis, which shows that the efficiency of a project that uses the FTSProc is greater than a project developed in an adhoc way, this section presents other factors that corroborate to this result.• Analyzing the qualitative results, we found that the advantage of the FTSProc lies in the fact that the FTSProc team realizes clearly and quickly how the work should be continued. Thus, the time for this identification is smaller than the adhoc team, resulting in a greater time for requirements development during each working day (shift). This advantage is due two main factors related to the FTSProc: usage of TDD and the main three questions that the process proposed. 24
  25. 25. Conclusions• This research presents a handoff process proposal to projects that use the FTS strategy.• The objectives of the proposed process in this work are focused on reducing the challenges posed by FTS strategy.• To evaluate the proposed process, an experiment was conducted, which demonstrated that the FTSProc actually achieves its goals, – alleviates the challenges present in projects that use the strategy FTS. 25
  26. 26. Contributions• Theory• The creation of a process for the work handoff during the development phase.• The process was proposed with data from the literature. After running the experiment and analyzed its results, we found that the process was indeed effective.• Since while the literature does not present a specific process for the development phase, this study proposed, performed an experiment, and pointed evidences of the effectiveness of the proposed process.26
  27. 27. Contributions• Market• Currently searching for competitive advantages such as cost reduction and productivity gains, companies are using offshore operations.• This work can contribute to increasing the productivity gain, since the created process facilitates the use of the FTS strategy for the development phase, thus decreasing the time spent during the this phase of SDLC. 27
  28. 28. Limitations• In the experimental process, due some schedule constraints, the number of participants able to join the experiment was low (8 people). Therefore, it was an impediment to use statistical methods to verify the hypothesis, opting then by a qualitative interpretation to analyze the results.• This interpretation presented evidences of the greater efficiency of the project that used the FTSProc, but not allowed to obtain conclusions with a significant confidence level, what could be achieve through experiments with statistical analysis of the results 28
  29. 29. Limitations• Specific generalization of the experiment due the fact that the scope of the project was fictitious and created by the researcher.• Application of an experimental research method, as the subjective influence of the researcher or the participants in the results. 29
  30. 30. Future Studies• With the purpose of substantiating the evidence presented by experimental method, as a future study, it is suggested the experiment replication with a larger number of participants to evaluate the FTSProc, which allows a significant statistical validation for obtaining conclusions on the hypotheses. 30
  31. 31. Future Studies• To carry out a study case to evaluate the use of the process created in a real environment;• Would be possible to verify the behavior of the process in a real environment• It would be possible to prove the results found in this study, through a controlled experiment, are equivalent in a real environment 31
  32. 32. Future Studies• Studies aiming to expand this process to other phases of the SDLC are relevant.• Other phases might be contemplated with a process to facilitate the FTS strategy usage.• At the end might be possible to create a process comprised by several sub-processes, which contemplate all phases of SDLC.• Thus, the entire software project could be accomplished using the FTS strategy and therefore, reducing the time of construction in all phases of a project. 32

×