Programme Specification
- 1. Programme Specification A statement of the knowledge, understanding and skills that underpin a taught programme of study leading to an award from the University of Sheffield 1 Programme Title Software Engineering 2 Programme Code COMU06 (MEng), COMU05 (BEng) 3 JACS Code G650 4 Level of Study Undergraduate 5a Final Qualification Master of Engineering (MEng) 5b QAA FHEQ Level Masters 6 Intermediate Qualification Bachelor of Engineering with Honours (BEng Hons) 7 Teaching Institution (if not Sheffield) Not applicable 8 Faculty Engineering 9 Department Computer Science Other Departments involved in 10 Pure Mathematics, Probability and Statistics teaching the programme 11 Mode of Attendance Full-time 12 Duration of the Programme Four years (MEng), Three years (BEng) Accrediting Professional or 13 British Computer Society Statutory Body 14 Date of production/revision October 2003 15. Background to the programme and subject area Software Engineering is the application of a systematic, disciplined and quantifiable approach to the development, operation and maintenance of software. Software systems now permeate every aspect of life and are amongst the most complex of manufactured artefacts; this requires skilled engineers to design and build them reliably and economically. A degree in Software Engineering covers a wide spectrum of knowledge, ranging from mathematics, algorithms and data structures, software engineering process and notations, hardware and networks to professional issues and business management. It also develops a wide range of technical and interpersonal skills, including analysis and design, computer programming, team management, report writing, presentation skills and enterprise. Our Software Engineering degree has a strong theoretical basis and a high practical content. Key features include project teamwork at all levels, building real business systems for external clients, a project dissertation and participation in the Genesys Solutions software company. At the higher levels, students may pursue advanced topics in software engineering, computer speech and language processing, 3D graphics and games, robotics and machine learning, theoretical computer science, distributed systems and e-commerce. Our curriculum reflects the latest research developments and professional standards in Software Engineering. The Department of Computer Science was ranked highly (grade 5) in the national Research Assessment Exercise. Close links are maintained with industry-leading companies such as IBM, EDS and HSBC. Our degrees are accredited by the British Computer Society, providing a route to the professional status of Chartered Engineer (CEng) and offering exemptions from BCS professional examinations. See the Department of Computer Science website: http://www.dcs.shef.ac.uk/ for more information. 16. Programme aims The aims of the programme are: 1. To provide a thorough academic grounding in the core subject matter of Software Engineering, with advanced study paths informed by the Department’s wide-ranging research interests; 1 programme-specification4402.doc – ver07
- 2. 2. To develop technical, professional and managerial skills through exposure to practical, industrially-oriented projects, emphasising teamwork and communication as well as software design and development skills; 3. To provide a route to professional accreditation through the British Computer Society, leading to partial (BEng) and full (MEng) CEng accreditation and exemptions from the relevant professional examinations; 4. To expose students to leading-edge world-class research in Software Engineering (BEng) and engage students in advanced research areas and methods (MEng); 5. To produce immediately employable graduates with an industrially relevant mix of knowledge, practical skills and self-motivation (BEng) and with leadership and enterprise skills (MEng); 6. To provide an international dimension with opportunities for exchanges with European, American, Canadian and Australian universities. 17. Programme learning outcomes Knowledge and understanding - On successful completion of the programme, students will have obtained: K1 A full understanding of programming languages and styles, algorithms and data structures. K2 A full understanding of discrete and continuous mathematical foundations for computing. K3 A full understanding of software engineering, analysis and design methods and process management. K4 A full understanding of artificial intelligence and biologically-inspired models of machine reasoning. K5 An appreciation of computer hardware design and computer network architectures. K6 An appreciation of human, social, legal and professional issues relating to the use of computers. K7 An appreciation of the commercial and industrial dimension to computing, through interaction with clients. K8 A deeper understanding of software quality and comparative methodologies (core topic for the degree). K9 A deeper understanding in two (BEng) to four (MEng) research-led advanced study paths in the areas of computer speech and hearing, natural language processing, distributed systems and e-commerce, 3D graphics and games, robotics and machine learning. K10 A broader understanding in up to two (BEng) or three (MEng) optional study paths in Psychology, Philosophy, Information Systems, Mathematics, Physics and Artificial Intelligence. K11 (MEng only): A deeper understanding of a selection of leading-edge advanced research topics offered annually by the Department of Computer Science’s research groups. Skills and other attributes - On successful completion of the programme, students will be able: S1 To function in an Information and Communication Technology (ICT) environment, using email, the internet and office software packages S2 To conceive, design and write correct working computer programs in several different programming styles, using a variety of compilers and development environments. S3 To construct and manipulate formal and mathematical models, use model-checking and mathematical modelling packages. S4 To apply a software engineering process and take a project through the stages of the software lifecycle, using design notations and software engineering tools selectively. S5 To communicate effectively in writing, present a two-sided argument, expose technical information clearly, comprehend and summarise research-level material with proper citation of sources. S6 To communicate effectively in speaking, interview and interact productively with a client, present and defend a substantial piece of work, engage with others and respond effectively to questions. S7 To work effectively in a team, demonstrating personal responsibility and group management ability, interpersonal skills, leadership and delegation, and plan to meet deadlines. S8 To develop industrially-relevant software systems for external clients in a competitive group environment, with incomplete and changing requirements, delivering to tight deadlines. S9 To research material from multiple published sources, comprehend and filter such material and from it synthesize theories, principles or designs pertinent to a practical, problem-solving project. S10 To demonstrate personal initiative, self-motivation and problem-solving skills, through the selection and taking through to completion of a practical, problem-solving individual project with a research dimension. S11 (MEng only): To demonstrate group initiative and enterprise within the Genesys Solutions software house, 2 programme-specification4402.doc – ver07
- 3. attracting and managing new company business with fee-paying clients from business and industry. 18. Teaching, learning and assessment Development of the learning outcomes is promoted through the following teaching and learning methods: The Department fosters an environment with many opportunities for individual and group learning, but the responsibility for learning rests with the student, who must be personally organised and self-motivated to make the most of the programme. Teaching is offered through formal lectures, seminars, computer laboratories, problem- solving classes and project supervision. Lectures are formal presentations to a large class of students by a lecturer. The purpose of a lecture is to motivate interest in a subject, to convey the core concepts and information content succinctly and to point students towards further sources of information. Lectures are interactive and students are encouraged to ask questions at suitable points. Students are expected to take notes during lectures, adding detail to published course materials (which should be printed and brought to the lecture, when provided in advance on electronic media). The learning outcomes K1-K10 are supported mainly through this mode. Seminars are longer semi-formal presentations to a class of students by a lecturer, researcher, industrial partner or student, describing an area of their current research or business. There is typically more opportunity than in a lecture to structure the session internally with questions, problem solving and other kinds of interactive or shared learning experience, in which the students may also participate in the teaching and lead discussions. The learning outcomes K7, K11 and S5-S9 are directly promoted through this mode, with indirect support for K1-K6, K8-K10. Computer laboratories are sessions supervised by teaching assistants (under the direction of the responsible lecturer) in which students work at a computer, to develop a specific practical skill, such as familiarisation, computer programming, or the use of a software engineering or mathematical modelling tool. The learning outcomes S1-S4 are promoted mainly through this mode, with indirect support for K1-K3. Problem-solving classes are sessions conducted by a lecturer with a class of students, in which exercises are completed interactively and solutions are provided within the period. The purpose of such a class is to help students engage with, and assimilate the material presented in lectures and start to apply this knowledge. The learning outcomes K2-K6 and K8 are supported through this mode. Project supervision involves regular meetings with a student’s individual or group project supervisor, who may also be their personal tutor. During each session, students report on their progress to the supervisor, who highlights further areas of investigation, helps with technical problems, advises about the content and structure of technical reports and generally encourages the students to organise their time effectively. The learning outcomes S5-S11 are directly promoted through this mode, with S1-S4 supported indirectly. The transition to self-motivated learning is encouraged through specialist teaching materials such as lecture handouts or copies of lecture slides, supplied via the Department’s website. Set course texts and background materials are available through the University libraries, at bookshops and also via the Internet. Active learning is fostered and promoted through engagement in practical work, such as exercises, assignments and projects. Exercises are short tasks, either writing computer programs or working out solutions to other kinds of set problem, which are typically reviewed at the end of the session. Learning outcomes K1-K10 and S1-S4 may be supported this way. Assignments are typically offered in stages over a number of weeks, involving the design and implementation of a software system to perform a given task, or the researching of a body of information leading to the writing of a discursive essay on a given topic. Learning outcomes S1-S5 and K11 are supported by this; indirectly K1-K10 are reinforced. Projects are undertaken individually or in groups over one or two semesters. Projects typically solve a larger problem, possibly for an industrial client, possibly with a research dimension. Individual projects require personal organisation and presentation skills; group projects also require group organisational and communication skills. Learning outcomes K6-K7 and S1-S11 are supported by this; indirectly, K1-K5, K8-K10 are reinforced. 3 programme-specification4402.doc – ver07
- 4. Private study makes up more than half of the time allocated to each module. Students are expected to read around the topics of each module and follow especially any directed reading from recommended course texts. Private study will include further investigations prior to exercises or projects and also consolidation of lecture notes. Opportunities to demonstrate achievement of the learning outcomes are provided through the following assessment methods: Modules may be assessed by formal examination, by practical assignments, by an individual or group project, or by some combination of these methods. Examinations are typically 2-hour question papers, in which students answer 3 from a choice of 4 questions. A typical question has 40% of the credit devoted to the recall of knowledge and information and 60% of the credit devoted to applying this knowledge actively to solve a short problem. Examinations test the knowledge learning outcomes K2-K10, but also provide evidence of practical skills S3 and S5, and, to a lesser extent, evidence of previous engagement in S2 and S4. Assignments are typically 10-20 hour pieces of continuously assessed coursework, which students complete individually or in groups as directed. An assignment may have multiple stages, each offered over a 2-3 week period, delivered to separate deadlines. Assignments both develop and assess the practical skills S2-S5 and they are the main means of assessing programming ability K1, S2. ICT skills S1 are assessed indirectly. Individual projects are completed at Level 3, typically over two semesters. Students select a topic, research the background literature, prepare a survey/analysis report at the interim assessment stage, and apply this knowledge in a practical, problem-solving project which typically involves the design, implementation and testing of a substantial piece of software. The final assessment stage is by dissertation and poster session, assessed independently by two examiners. A viva voce examination may be held to form a common view in cases of insufficient evidence or divergent opinions. The learning outcomes S5-S6, S9-S10 are directly assessed, together with specialist areas of knowledge from K8-K9. Practical skills in S2-S4 and knowledge in K1-K7 may be assessed indirectly. Group projects are completed at Levels 1, 2 and 4, over one or two semesters. Student teams are given topics (Level 1) or negotiate topics with their industrial clients (Levels 2, 4). Teams prepare analysis and design documents, conduct client interviews (Levels 2, 4), demonstrate a tested, working software system, which may also involve research into designs or theories (Level 4), and provide a final report, together with timesheets, minutes and other evidence of their group management strategy. Credit is awarded to the team as a whole on the basis of the quality of the work, as evidenced in the final report, interim documentation (all Levels) and reported client satisfaction (Levels 2, 4). Credit is weighted towards individual team members based on their participation, as evidenced in the minutes, timesheets and other indicators of the division of workload and responsibility, which may include viva voce interviews. The learning outcomes S2, S4-S8 are directly assessed in software projects, together with K6-K7 in industrial projects, and K11, S5-S7, S9 and S11 in projects with a research dimension, which may also assess S2- S4 according to the type of project. Indirectly, K1-5 S1 may be assessed (all Levels) and K8-K10 (Level 4). Relative proportions of types of assessment by level A range of values indicates variation due to elective module choices Examination Assignment Individual Project Group Project Level 1 30-47% 37-54% 0% 16% Level 2 48% 36% 0% 16% Level 3 17-50% 0-33% 50% 0% Level 4 0-33% 34-67% 0% 33% 4 programme-specification4402.doc – ver07
- 5. 19. Reference points The learning outcomes have been developed to reflect the following points of reference: • The curriculum conforms to the UK Subject Benchmark in Computing, 2002. • The dual pathway structure through the degree programme, leading to separate MEng and BEng outcomes, was determined by SARTOR, 3rd Edition, 1997, which is considered the primary point of reference for degrees offering professional accreditation. • The QAA Framework for Higher Education Qualifications, 2001 is a further point of reference. • The Software Engineering degree programme is formally accredited by the British Computer Society, through whom conformance to SARTOR is regulated. The prominence given to practical, industrially related project work is supported by the BCS and our Industrial Liaison Board, which represents leading companies such as IBM, HSBC and EDS. • Mandatory parts of the core curriculum at Levels 1-2 were determined in consultation with the visiting Accreditation Panel from the British Computer Society in November 2000, which also certified the depth and breadth offered at Levels 3-4, including the capstone individual and group projects. • The advanced study paths at Levels 3-4 were devised in consultation with peer reviewers from Imperial College and York Universities and designed to fit the future requirements for Further Study under the UK SPEC guidelines (formerly, Matching Sections). • Their content is directly informed by the research interests of the Department and so conforms to the University Mission Statement to provide research-led teaching. • The workload fits comfortably within the guidelines laid down by the University, and is monitored by external examiners, who also review the content and standards of the programme. 20. Programme structure and regulations Our degree programmes are designed with a common curriculum at Levels 1-2, broadening out into many advanced study paths at Levels 3-4. The common core satisfies the requirements for accreditation by the British Computer Society. It also permits direct transfers between the related degrees in Computer Science, Software Engineering and Artificial Intelligence and Computer Science and delayed transfers from related dual honours programmes. The advanced level curriculum is designed around subject threads. Each thread is a suggested study path in a particular specialism, consisting typically of paired honours and masters modules, optionally supplemented by a related unit from the Advanced Research Topics programme (according to availability). The Software Engineering degree is distinctly characterised by the Software Engineering thread at Level 3, and participation in the Genesys Solutions company at Level 4. Apart from this, students may select advanced subject threads that match their interests or complement their chosen project topic. Level 1 has a fixed core of 100 credits, and 20 unrestricted credits, offering a free choice of subsidiary modules. The core consists of five subject threads running through the year: Introductory Programming, Software Engineering, Mathematical Foundations, Artificial Intelligence and Computer/Network Architectures. The Crossover Project, a group systems development exercise, is an integral part of the Software Engineering thread. Level 2 contributes to the final degree classification (1/4 BEng, 1/7 MEng) and has a completely fixed structure, consisting of a 100-credit core and a 20-credit group project, which is mandatory for the Software Engineering degree. The core consists of five subject threads: Programming and Algorithms, Intermediate Mathematical Theory, Industrial Software Engineering, Applied Artificial Intelligence and Human/Professional Issues. The Software Hut, an industrial group project developing software systems for external clients, is taken in semester 2. Students may transfer freely between this programme and the single honours degree programmes listed above at any time up until the end of Level 2. Dual honours students may transfer back into the single honours programme at the previous Level. Transfers between MEng and BEng are also freely permitted up until the end of Level 2, at which point students must maintain a higher standard to continue on the MEng, otherwise they must transfer to the BEng. Students must satisfy the prerequisites for the Software Engineering thread at the end of Level 2 to remain on the Software Engineering degree, otherwise they are advised to transfer to one of the related degrees. Resit examinations are held in August for Levels 1 and 2, though project-based assessments and some coursework cannot be repeated within the same year. 5 programme-specification4402.doc – ver07
- 6. Level 3 contributes to the final degree classification (3/4 BEng, 3/7 MEng) and consists of the core 60-credit Individual Project, approved units to the value of 40 credits, which must include at least 10 credits from the Software Engineering thread, and a further 20 unrestricted credits in a related discipline. The approved threads for this degree are: Software Engineering (mandatory), Natural Language Processing, Speech and Hearing, Computer Graphics and Games, Distributed Systems and E-Commerce, Machine Learning and Theoretical Computer Science. The unrestricted credits are typically taken in Information Studies, Psychology or Philosophy, but may also include modules from the Adaptive Intelligence thread and threads not taken above. BEng students graduate on completing Level 3 successfully (see full regulations). MEng students must maintain a higher standard (no lower than class 2/ii) to proceed to Level 4, otherwise they may seek to graduate immediately under the BEng regulations. The Individual Project must be passed at the first attempt to gain honours and accreditation (see section 21). A student failing to graduate on the first attempt may repeat Level 3 once and hopefully graduate, but without honours. Level 4 contributes to the final degree classification (3/7 MEng) and has a 60-credit core, consisting of the 40-credit group project in Genesys Solutions, and 20 credits taken from the Advanced Research Topics (ART). Approved units to the value of 40 credits are taken, typically from the advanced subject threads (see above), but which may include 20 further credits from the ART programme. A further 20 unrestricted credits may be taken in a related discipline (see above). The ART programme is a showcase for leading-edge research and a selection of short ART units are offered annually by the Department’s research groups (see section 23), allowing students to build a portfolio of preferred topics. Genesys Solutions offers a professional company context in which to develop enterprise skills. MEng students graduate on completing Level 4 successfully (see full regulations below). A student who, exceptionally, fails to meet the required standard (no lower than class 2/ii) may graduate with a BEng. Detailed information about the structure of programmes, regulations concerning assessment and progression and descriptions of individual modules are published in the University Calendar available on-line at http://www.shef.ac.uk/ govern/calendar/regs.html. 21. Student development over the course of study Level 1: Students learn the Java programming language, covering the basics of syntax, program construction and compiler tools. Later, they learn about object-oriented design, standard library packages, and how to use Java’s self- documentation. Students learn about software lifecycles, the Unified Modelling Language (UML) design notation and the use of various Computer-Aided Software Engineering (CASE) tools. Through the Crossover Project, student teams learn how to pick up projects partway through the software lifecycle and progress them to the next stage. Students also acquire discrete mathematics for logic and formal specification, followed by matrix manipulation, probability and statistics, aided by the MatLab tool. Students learn about human and machine intelligence, and are trained in how to present a reasoned argument. Later, they learn how to program intelligent search algorithms in Java. Students learn about different kinds of network and the OSI reference model for network protocols. Later, they learn about logic circuits in computer hardware and how to program simple hardware devices in Java. Level 2: Students learn the Haskell programming language, and the different functional style of programming using recursion and lazy evaluation. Later, they develop skills in problem abstraction, designing algorithms around abstract datatypes in Haskell. Students learn about automata, formal languages and the theory of computation, followed by further matrix manipulation and mathematical transforms for pattern recognition. Students learn database technology and are trained in a complete systems analysis and design method, using design notations learned at Level 1. Later, student teams compete in the Software Hut industrial project to deliver systems for external clients, applying the techniques and methods learned earlier. Students use a simulator to train a robot to avoid obstacles, using neural and evolutionary algorithms. They also learn the Prolog programming language, and the different logical style of programming using pattern-directed search. Students learn about legal, social and ethical issues in relation to the use of computers; and also the principles of human/computer interface (HCI) design, with further practical Java experience in building graphical user interfaces (GUIs). Level 3: The capstone achievement is the Individual Project, a major piece of research and development requiring initiative and problem-solving skills, which occupies fully half of a student’s time during the year. Literature research leads to a survey and analysis report in the Autumn, followed by detailed design, implementation and testing, and a written dissertation in the Spring. Students demonstrate their working systems at a poster session, run like a technology fair with industrial visitors. Other knowledge and skills depend on the particular subject threads followed. Software Engineering (mandatory) includes software measurement and testing methods, and comparative analysis of software engineering methodologies. Natural Language Processing includes text manipulation, information 6 programme-specification4402.doc – ver07
- 7. extraction, parsing and semantic interpretation. Speech and Hearing includes models of speech production and hearing, automatic speech recognition and synthesis. Computer Graphics and Games includes 3D modelling, rendering, collision detection, animation and intelligent behaviour. Distributed Systems and E-Commerce includes network performance analysis, distributed systems programming in Java and developing a web-based e-commerce system with a back-end database. Machine Learning includes symbolic, neural and evolutionary models of learning, information theory and its applications. Theoretical Computer Science includes symbolic calculi, model checking, types and program semantics. Level 4: The capstone achievement is the group enterprise project, taken in Genesys Solutions, which occupies a third of a student’s time during the year. In both of these, students learn to organise themselves professionally, holding business meetings and producing action minutes. The groups act on their own initiative to determine strategy, make decisions, allocate resources and evaluate progress. In Genesys Solutions, the group attracts new business from external fee-paying clients in industry and the public sector, develops new products and maintains the company’s software base. Other knowledge and skills depend on the particular additional subject threads followed (see above), and the particular topics chosen on the Advanced Research Topics programme, which vary from year to year (see section 23). Upon Graduation: Students successfully completing the 3-year BEng programme graduate with the title: Bachelor of Engineering with Honours (BEng Hons) in Software Engineering. Students successfully completing the 4-year MEng programme graduate with Master of Engineering (MEng) in Software Engineering. Professional Development: The Software Engineering degree programme (MEng, BEng) is accredited by the British Computer Society (BCS), the Chartered Institute for Computing and Information Systems. Accreditation recognises that a degree programme meets the requirement for the professional formation of a Chartered Engineer. The BCS also grants exemptions from its own examinations. Accreditation opens the way to the professional status of Chartered Engineer (CEng). Exemptions provide a route to full membership of the BCS (MBCS, FBCS). Students graduating with BEng Hons. In Software Engineering automatically obtain Partial CEng Accreditation and are granted exemptions from the following BCS examinations: the Certificate, the Diploma and the Diploma Project. Students graduating with MEng Hons. in Software Engineering automatically obtain Full CEng Accreditation and are granted exemptions from all the above BCS examinations, plus the following: the Professional Graduate Diploma and the Professional Graduate Diploma Project. After gaining at least three years’ professional work experience in the computing and information systems field, MEng graduates may apply to the British Computer Society to obtain the status of Chartered Engineer. BEng graduates who subsequently complete a further advanced study programme (such as an accredited MSc in Software Engineering or Computer Science) may likewise apply after gaining similar professional work experience. 22. Criteria for admission to the programme Detailed information regarding admission to the programme is available at http://www.shef.ac.uk/prospective/ The standard requirement is three A-levels, of which one must be in Mathematics. Other equivalent national and international qualifications are accepted; see http://www.dcs.shef.ac.uk/ for more details. Wider access is also offered to students lacking suitable A-level qualifications for science and engineering through the Foundation Year; see http://www.shef.ac.uk/foundation/ for more details. A general University of Sheffield admissions requirement is GCSE English. Non-native speakers of English must demonstrate suitable competence in the language. The preferred test is IELTS, though others are accepted. See the English Language Teaching Centre website http://www.shef.ac.uk/eltc/ for more information. Students wishing to participate in the Socrates/Erasmus European exchange programme should have obtained at least a GCSE in the foreign language of their chosen host country. 23. Additional information The Department of Computer Science is housed in the modern, purpose-built Regent Court building and has its own dedicated computing facilities. The Department is internationally recognized for its teaching and research (grade 5 in the last Research Assessment Exercise), and has particular research strengths in the fields of verification and testing, natural language processing, speech technology, computational biology, machine learning, robotics and 7 programme-specification4402.doc – ver07
- 8. computer graphics. Its project-led teaching has been copied at other universities and the enterprise culture promoted in Genesys Solutions was featured as an example of excellence by the Royal Academy of Engineering and attracted the personal endorsement of the Secretary of State for Education. An international dimension is offered through the Socrates/Erasmus European exchange programme, in which students may exchange their Level 2 year for an equivalent year at one of 5 universities in France and Germany; and through the Year Abroad exchange programme, in which students may likewise spend a year in America, Canada or Australia. See http://www.shef.ac.uk/international/ for more details. Furthermore, MEng students specialising in Speech, Hearing and Natural Language may elect to spend one semester of Level 4 at a European university participating in the EU programme: European Masters in Language and Speech. Satisfactory completion of an agreed study programme at both home and foreign institutions will gain the additional award of a European Masters degree (dual certification). See the web site at http://www.cstr.ed.ac.uk/euromasters/ for more information. The Department offers course materials and information services to students via the departmental intranet. The Department of Computer Science Undergraduate Student Handbook governs all local aspects of academic student life, with regard to services offered, computer etiquette, and local regulations. See the Department of Computer Science website: http://www.dcs.shef.ac.uk/ for more information. This specification represents a concise statement about the main features of the programme and should be considered alongside other sources of information provided by the teaching department(s) and the University. In addition to programme specific information, further information about studying at The University of Sheffield can be accessed via our Student Services web site at http://www.shef.ac.uk/ssid/. 8 programme-specification4402.doc – ver07