BSc: Lean Software Development

Lean Software Development

• Course name: Lean Software Development
• Code discipline: XYZ
• Subject area:

Short Description

This course covers the following concepts: Fundamental principles of producing software as a creative act of the human mind; Techniques to optimize such production, with specific focus on agile methods.

Course Topics

Course Sections and Topics

Section	Topics within the section
Software as a creative activity	Nature of software Software and art Core resources for the production of software Tame and wicked projects Organizing the activities based on the GQM approach
Measurement in software	Meaning of measures The representational theory of measurement Measurement scales Fundamental measures for the production of software Procedural measures Object Oriented measures
Taylorism and Fordism	The increase of productivity in the idea of Taylor The role of division of work Planning and formalization of tasks Economies of scale Problems in understanding tasks Taylorism/Fordism and software development
Lean and Agile	Taiichi Ono and the Toyota Production System Creating a “Radiography” of the Production Process Workers involvement “Pull” and Not “Push” Kanban Quality management Process control Job enrichment Control and coordination mechanisms Case study: Extreme Programming
Issues in Lean and Agile	The “Hype of Agile” The dark side of agile Skepticism about agile methods Knowledge and software engineering Using burn-down charts The Zen of agile
Structuring a Lean Approach to software development	Existing proposals to create a “Lean Software Development” Sharing a common vision Depriving gurus of their power GQM+ Applying the GQM+ step-by-step Business alignment GQM+ for business alignment
Optimizing the development process	Why the PDSA does not work in software The experience factory The QIP cycle Non invasive measurement The big-brother effect The role of autonomation Employing Andon boards

Intended Learning Outcomes (ILOs)

What is the main purpose of this course?

This course exposes the student to the core concepts behind Lean Development in Software Engineering, beyond myths and legends, emphasizing how it relates to the general principles of Lean Development. It discusses the different possible software processes, how they can be tailored, enacted, and measured. In addition, a significant part of the course is centered around the application of lean to software development to knowledge intensive areas not necessarily connected to software.

ILOs defined at three levels

Level 1: What concepts should a student know/remember/explain?

By the end of the course, the students should be able to ...

• creative nature of software production as an act of creativity of the human mind,
• the substantial differences between tame and wicked problems,
• the core concepts of measurement in software engineering,
• the fundamentals of Taylorist/Fordist approaches to (software) production,
• the basis of lean and agile software development,
• the “dark” side of agility,
• the importance of knowledge and knowledge sharing in producing software,
• how to create an ad-hoc process for a development organization.

Level 2: What basic practical skills should a student be able to perform?

By the end of the course, the students should be able to ...

• when a problem is “easy to solve” provided enough effort is put to such solution and when it is not,
• what is a measure in general,
• why it is important and how we can define and perform measurements in software engineering, especially in lean and agile development environments,
• how to organize the development process to collect metrics non invasively,
• the difference between pulling and pushing in (software) development,
• the fundamental principle of agility,
• the risk intrinsic in the dark side of agile,
• how to organize an agile development process based on the definition of overall Goals, associated Question, and milestones based on Metrics,
• an environment based on experience, like the Experience Factory.

Level 3: What complex comprehensive skills should a student be able to apply in real-life scenarios?

By the end of the course, the students should be able to ...

• Compute the fundamental software metrics to track the evolution of a project,
• Organize the aims of a (software) development organization in terms of Goals, Questions, and Metrics,
• Create a tailored (lean and agile) development process for an organization producing software,
• Define a path to insert and manage such (lean and agile) development process into an organization producing software,
• Structure the experience gathering during inside an organization to based on it the future strategic decision of such organization,
• Relate the various proposals for Agile Methods to the overall principles of Lean Management,
• Define a suitable (lean) process for a new organization, a process to introduce and institutionalize it, and an approach to measure the outcome of such introduction and institutionalization.

Grading

Course grading range

Grade	Range	Description of performance
A. Excellent	95-100	-
B. Good	80-94	-
C. Satisfactory	51-79	-
D. Poor	0-50	-

Course activities and grading breakdown

Activity Type	Percentage of the overall course grade
Class and lab participation (including class quizzes)	25
Intermediate mini-oral	25
Final Oral Exam	25
Project	25

In particular:

• Every component apart from class and lab participation carries up to 30% extra credit.
• Students aiming for a C, have only to pass Class and lab participation and the intermediate mini-oral (after week 2).
• Students aiming for a B also have to sustain the final oral exam.
• Students aiming for an A also have to do the project.
• Failing any attempted component of the course implies failing the entire course and go to the retake.
• The final grade is computed summing the points taken in each part and considering the table above, so it could be that, for instance, a student aiming for an A but not getting enough points in each of the component would then get a C or a B.

For this reason it is strongly recommended that each student, based on her/his desire, load, condition, etc decides at the beginning of the course which grade to aim at.

Recommendations for students on how to succeed in the course

The goal of the project is to read the book and understand what is the takeaway for software engineering, considering three fundamental aspects: a) the process to follow, b) the product being built c) the structure of the team and its organization. The projects will be executed in weekly iteration on traceable files.

The projects will be partially graded weekly (30%) and part at the end (70%). Moreover, at the end you will need to give a 5-minute presentation of your work.

During beginning of the project you will create the overall GQM of the project and you will be initially evaluated based on it. After setting the GQM, you will also create the vision of your project with a roadmap. Based on the roadmap, you will start to have the reviews of your project increments in terms of the a) progresses toward the completion b) accurate planning based on the defined GQM, and c) quality of the work. Weekly grade ranges from 0 to 2 points.

The overall project success will be evaluated as following:

Course activities and grading breakdown

Grade	Project outcome
A (100%)	In-depth analysis of the subject under investigation with original ideas for SE and practical experiments
B (80%)	In-depth analysis of the subject under investigation with original ideas for SE
C (60%)	Analysis of the subject under investigation, highlighting general ideas for SE, without novelty
D (0%)	No activities

Resources, literature and reference materials

Open access resources

• Textbook: Andrea Janes and Giancarlo Succi. Lean Software Development in Action. Springer, Heidelberg, Germany, 2014. ISBN 978-3-662-44178-7. doi: 10.1007/978-3-642-00503-9.
• Reference: James P. Womack and Daniel T. Jones. Lean Thinking: banish waste and create wealth in your corporation. Lean Enterprise Institute. Simon & Schuster, 1996. ISBN 9780684810355.
• Reference: Taiichi Ohno. Toyota production system: beyond large-scale production. CRC Press, 1988
• Reference: James P. Womack. Lean Thinking. Simon & Schuster, Limited, 1997. ISBN 9780671004712.
• Reference: James P. Womack, Daniel T. Jones, and Daniel Roos. The Machine That Changed the World: The Story of Lean Production. Harper Perennial modern classics. HarperCollins, 1991. ISBN 9780060974176.
• Reference: Mary Poppendieck and Tom Poppendieck. Implementing Lean Software Development: From Concept to Cash. A Kent Beck signature book. Addison-Wesley, 2007. ISBN 9780321437389.
• Reference: C.M. Christensen. The Innovator’s Dilemma: When New Technologies Cause Great Firms to Fail. Management of innovation and change series. Harvard Business School Press, 1997. ISBN 9780875845852.

Software and tools used within the course

Moodle, Miro, Overleaf

Teaching Methodology: Methods, techniques, & activities

Activities and Teaching Methods

Activities within each section

Learning Activities	Section 1	Section 2	Section 3	Section 4	Section 5	Section 6	Section 7
Homework and group projects	1	1	1	1	1	1	1
Reports	1	1	1	1	1	1	1
Oral polls	1	1	1	1	1	1	1
Discussions	1	1	1	1	1	1	1

Formative Assessment and Course Activities

Ongoing performance assessment

Section 1

Activity Type	Content	Is Graded?
Question	Provide examples of creativity in the production of software.	1
Question	Describe the differences between tame and wicked projects.	1
Question	Discuss the key resources needed for the production of software.	1
Question	What are key issues in creative production of software for distributed teams?	1
Question	Provide examples of wicked problems from your everyday life	0
Question	Evidence wickedness in different aspects of software production	0
Question	Create a GQM for your aims of the semester	0
Question	Discuss the role of GQM in tame and wicked projects	0

Section 2

Activity Type	Content	Is Graded?
Question	Provide examples of the representational theory of measurement	1
Question	List the measurement scales	1
Question	Present for each measurement scale the operations that can be performed on it	1
Question	Discuss the representational condition	1
Question	What are key size metrics?	1
Question	What are key complexity metrics?	1
Question	Provide examples of subjective and objective metrics	0
Question	List 3 direct and 3 indirect measures, evidencing also the problems connected to the construction of indirect measures	0
Question	Compute LOC and other size metrics for code snippets	0
Question	Compute MCC and other complexity metrics for code snippets	0
Question	Compute the metrics of the CK suite for portions of Object Oriented systems	0

Section 3

Activity Type	Content	Is Graded?
Question	Detail the fundamental assumptions by Taylor	1
Question	What are the fundamental activities of managers according to Taylor?	1
Question	In which sense Taylor has influenced what we now consider “good management practices?”	1
Question	How does creativity relates to the “good management practices” of Taylor?	1
Question	What are the problems in applying Fordism/Taylorism to software development?	1
Question	Provide examples of companies where the approach by Taylor has been successful	0
Question	Discuss how the approach of Taylor can be useful in attracting and/or retaining employees	0
Question	Analyse the fundamental activities of managers according to Taylor and determine their limits	0
Question	Determine the fundamental activities of managers according to Taylor can have an implication in the software crisis	0
Question	Outline how flexibility and variable requirements can be handled in the context of Taylorism and Fordism	0

Section 4

Activity Type	Content	Is Graded?
Question	Present the key problems in batch production	1
Question	Outline the key principles of the approach of Ono at Toyota	1
Question	What are the fundamental steps in eliminating waste according to Ono?	1
Question	Details the role of the customers and of the workers in the approach of Ono	1
Question	Explain the difference between “Pulling” and “Pushing”	1
Question	What are key steps in improving quality according to Ono?	1
Question	What are the key control and coordination mechanisms available?	1
Question	What are the fundamental two actions needed to perform a Lean transformation according to Ono?	0
Question	What are the associated three major needs?	0
Question	What are the 5 steps to enact Lean Thinking according to Womak and Jones?	0
Question	Discuss the 8 constantly ongoing activities in a lean company like Toyota	0
Question	How can activities been classified in a decision matrix in an environment like Toyota?	0
Question	Provide concrete examples of “Push” and of “Pull” in software production.	0
Question	Details the control and coordination mechanisms present in agile and in traditional development environments.	0

Section 5

Activity Type	Content	Is Graded?
Question	What is the Gartner’s Innovation Hype Cycle? Could you provide examples of application of it to a different field of knowledge?	1
Question	Describe the so-called “Dark Agile Manifesto”	1
Question	What are the sources of the skepticism present with respect to Agile?	1
Question	What makes agile awkward in the eyes of “traditional” managers?	1
Question	Given a production system, how do you determine what is the value of every step, how much improvement can be considered enough, and when is the point reached where value is not increased anymore but destroyed?	1
Question	How is it possible to obtain knowledge about the production process? How can I create visibility of the ongoing activities or problems?	1
Question	Identify in other areas of software engineering phenomenon similar to the “Dark Side of Agile.”	0
Question	Identify in other knowledge-intensive fields phenomena similar to the “Dark Side of Agile” and discuss how they can be tackled.	0
Question	Given a production process, determine strategies to store knowledge to create experience?	0
Question	How can you design the production process so that the team uses the gained experience?	0
Question	How can you systematically improve your process, also building on the experience to anticipate problems (create wisdom)?	0

Section 6

Activity Type	Content	Is Graded?
Question	Reflect on the seven principles for characterizing Lean Software Development by Mary and Tom Poppendieck.	1
Question	What is the semantic gap and which threats it poses to effective software development?	1
Question	Map the structure of Extreme Programming to the layered structure of Shalloway et al.	1
Question	How does the scientific method deprives gurus from their power?	1
Question	What are the steps to implement the GQM+?	1
Question	For what reason measurement goals and business goals should be interconnected?	1
Question	How do the practices of Lean Management defined by Hibbs and colleagues relate to the seven principles by Mary and Tom Poppendieck	0
Question	Propose how you could develop software and hardware tools to promote common visions in companies.	0
Question	Elaborate a proposal to create some kind of Balanced Scorecards to evaluate your current study.	0
Question	Discuss the opinion of Ono about following plans and the extent to which such opinion contradicts (a) the practices, and (b) the principles of the tayloristic/fordistic approach	0
Question	Propose a SWAT analysis to introduce a Lean approach to your most recent software development endavour	0

Section 7

Activity Type	Content	Is Graded?
Question	Details the major components of an experience factory.	1
Question	What are Reflection, Retrospective, and Post-Mortem Analysis? Why are they useful in Lean Software Development?	1
Question	What are the key components of a non invasive software measurement systems?	1
Question	What is the big brother effect and how it is possible to alleviate it.	1
Question	Discuss how autonomation is present in Extreme Programming.	1
Question	Where is the term “Dashboard” coming from and what is its use in Lean Software Development?	1
Question	List the steps of a QIP.	0
Question	Discuss the risks of a measurement program and how non invasive software measurement can help alleviating them.	0
Question	Are there cases in which Theory X of management could be more effective than Theory Y? Discuss your findings.	0
Question	For which aspects of software production autonomation could be useful?	0
Question	Which tools could be used to promote autonomation?	0
Question	Prototype by paper and pencil an Andon board that you would consider useful in a software production environment.	0

Final assessment

Section 1

1. Present the salient aspects of wicked problems
2. List the key aspects of software that make it a wicked problem
3. Provide a link between wickedness and creativity in software production
4. Discuss what promotes and what inhibit creativity in general and in software production
5. Outline meaning and limitation of the concept of “engineering” the production of software

Section 2

1. Structure the aims of a company using the GQM and detailing the metrics to compute, also explaining the deductions and the predictions that can be made with such metrics.
2. Given an index computed as a combination of metrics, determine if it is a metrics according to the representational theory of measurement.
3. Analyse a portion of a system and determine suitable metrics to extract and the information that would be provided by such metrics.
4. Given a website performing a service (like flight reservations), compute the Function Points for such website.
5. Discuss how to structure a taxonomy of quality for a specific company, explaining the role of reliability in it, and detailing how to compute the reliability again in the context of such company; please make the assumptions that you need to perform such computation.

Section 3

1. What aspects of making software feel like an art, which like a craft to you and how could you bend Taylorism and Fordism to handle it?
2. How does Fordism and Taylorism can explain that companies producing software containing bugs can still stay in the market, and, in certain case, also be successful?
3. Based on your experience and previous courses, which development models can refer to Fordism and Taylorism and which cannot be reduced to it?
4. How does the specialization of work in software development can be linked to Fordism and Taylorism?
5. Provide an example for each of the fundamental activities of managers according to Taylor that shows how such activity is very useful in software development and one that shows that is is inadequate.

Section 4

1. Structure a model like PDSA for a company producing websites for online marketing.
2. How can the right part and the right information be always available without waste according to the Toyota approach?
3. Compare a street crossing based on traffic lights with a roundabout and determine the approach that is safest and the one with the a highest throughput according to Waterfall and to Lean.
4. Discuss the involvements of workers in tayloristic/fordistic and in Lean development processes and their implications for the retention and the improvement of the quality of the workforce.
5. Outline the extent of which economies of scale exist in Lean development processes.
6. Imagine you had to introduce Extreme Programming in a software development team that follows a waterfall process. Which problems do you foresee? How will the clients react (that until now are used to work with a team that used the waterfall process)? How would you address them?

Section 5

1. Imagine you are the boss of a small software development company. Which actions would you do or which practices would you introduce to prevent that your programmers fall into the trap of following a software guru?
2. Elaborate possible extreme (and damaging) positions that can be taken by gurus of agile.
3. Explain the differences in introducing a methodology by a guru and by a smart and effective coach.
4. Detail why Extreme Programming produces an informative workspace.
5. Discuss effective ways of packaging and “distributing” knowledge in software teams, starting with the guru approach of organizing knowledge into simple, clear practices which are easy to explain and to follow.

Section 6

1. Discuss the role of customer on-site under the perspective of lean and outline it relevance in the earlier and then in the later proposals of agile software development.
2. Suppose that you have to develop Balanced Scorecards for a software development team. Which perspectives would you use? Which goals would you use for each perspective?
3. What is a socio-technical system and how can it be used to describe an (agile) software production environement?
4. Imagine you want to evaluate how readable the source code of some program is. Define a GQM+ model to describe what and why you would measure.
5. The development in company M occurs according to the following schema: when a new project is started, a developer takes an old project that is the most similar to the new requirements and makes a copy and starts implementing the required modifications. To improve this process and to help the company to adopt a component-based approach, we want to understand which pieces of code are the best candidates for future components and which variability points they have. Define a suitable GQM+ for such purpose.

Section 7

1. What type of wisdom (in the sense of “know-why”) would you manage in an Experience Factory to support Lean Thinking? Distinguish between organizational learning and project learning.
2. Assume you are a manager convinced that Theory X is true. Which non-invasive measurement probes would you want to develop to maximize productivity? Now assume you are convinced that Theory Y is true. Which non-invasive measurement probes would you need now?
3. We discussed that we foresee two ways to collect measurements non- invasively: in batch and in background mode. What are the advantages and disadvantages of each approach?
4. There are many interrelated building blocks (or concepts) of Lean Software Development and each contributes differently to it. What types of data are handled by each of these building blocks (or concepts)? How do the contribution to the overall value stream, to the creation of knowledge, and to the overall improvement?
5. Assume you set up a fantastic dashboard for your team. As you collect the data and visualize it, you notice that all the measurements show problematic values. You let the dashboard in place for some days and also show it to your collaborators, but nobody cares; everybody continues his job as if everything would be fine. What is going wrong?