Assignment 1 1625 Managing a Successful Computing Project đại học Greenwich 2022, điểm chuẩn Pass. Project initialization, Main aim of project, List of Objectives to achieve the aim, Project management plan, Scope, Time, Gantt chart, Communication, Resources, Risks, Cost estimation, WBS, Primary research, List of interview question, Summary about interview , List of survey question , Summary about survey, Secondary research, By 2025, the amount of data will reach 175ZB, which is a huge number. Imagine the computer hard drive is only 1TB then the difference will be obvious. As data increases, the number of terminals such as hard drives and phones also increases, which consumes a lot of power as well as resources, which is a reason for environmental pollution. The main reason for the increase in data is that the old data is still stored and the data that is not needed or is no longer useful is kept.
ASSIGNMENT FRONT SHEET Qualification BTEC Level HND Diploma in Computing Unit number and title Unit 06: Managing a Successful Computing Project Submission date 29/10/2022 Date Received 1st submission 29/10/2022 Re-submission Date 01/11/2022 Date Received 2nd submission 03/11/2022 Student Name Nguyen Manh Tung Student ID GCH200064 Class GCH0907 Assessor name Nguyen The Lam Tung Student declaration I certify that the assignment submission is entirely my own work and I fully understand the consequences of plagiarism I understand that making a false declaration is a form of malpractice Student’s signature Grading grid P1 P2 P3 P4 M1 M2 D1 Summative Feedback: Resubmission Feedback: 2.1 Grade: IV Signature: Assessor Signature: Date: Contents Introduction I II Project initialization 1 Main aim of project List of Objectives to achieve the aim III Project management plan Scope 2 Time 3 Communication Risks Resources Cost estimation Planning WBS Gantt chart Primary research List of interview question Summary about interview List of survey question Summary about survey 10 Evaluation 10 Secondary research 10 List of articles/ books 11 Summarize 11 Evaluation 13 Figure 1: WBS Figure 2: Gantt chart Figure 3: Suvery question (1) Figure 4: Survey question (2) Figure 5: Survey question (3) Figure 6: Survey question (4) Figure 7: Survey question (5) 10 I Introduction Problem: By 2025, the amount of data will reach 175ZB, which is a huge number Imagine the computer hard drive is only 1TB then the difference will be obvious As data increases, the number of terminals such as hard drives and phones also increases, which consumes a lot of power as well as resources, which is a reason for environmental pollution The main reason for the increase in data is that the old data is still stored and the data that is not needed or is no longer useful is kept Solution: Stemming from the above reasons, the proposed solution will focus on the input data, process it in real time, filter and store the necessary data, and delete the data that is no longer effective To implement this solution, AI and IoT are necessary tools IoT will collect real-time information, process it first if it's not too complicated, then send it back to the central processor, AI will decide what data is data back, what actions are taken, etc Project: As a member of the research and development department of the National Hydrometeorological Administration, I will be working on a project called "Climate Management AI" With the characteristics of the meteorological industry, the data sent every day is very large such as rainfall, air humidity, satellite images, tropical depressions, etc The amount of data to be processed is very large, a lot of data is not important or often encountered causing time consuming Due to the large amount of data, manual deletion takes a long time, possibly deleting important data by mistake Large capacity leads to greater power consumption to run This project will integrate AI to solve the above problem II Project initialization Main aim of project Integrate data processing technology for the system Input data processing eliminates redundant data Reduce newly generated data Reduce the workload for employees More accurate prediction of phenomena Easy data statistics List of Objectives to achieve the aim Learn about data types (work with data researchers) 1.1 Collect commonly sent weather data types 1.2 Learn about their influence on the weather Learn similar systems (hospital applications, etc.) 2.1 Working with agencies using the system 2.2 Find out advantages, disadvantages, points to note Understanding AI technology for systems (Machine learning) Building AI for the system (data processing and weather forecasting system) 4.1 Requirements 4.2 Analysis and design 4.3 Deployment (Encoding) 4.4 Testing Evaluation 5.1 Get a technical staff assessment 5.2 Provide general assessment and reporting 5.3 Provide direction for development III Project management plan Scope The "Climate Management AI" project will be completed within months This project is an AI with functions such as collecting and processing weather data, cleaning up redundant data, it is developed and applied to the current weather management system During the first week, we will analyze the feasibility, identify the risk and plan for it The technology applied will also be explored along with the study of the climate center's actual data Finally, make a specific plan for the whole project In the next weeks, work related to building the system will be carried out, including user requirements analysis (station staff), system design We spent weeks programming the system and applying it to the existing system We spend the next weeks testing in the climate center to be able to give the most accurate assessments In the last week, assessments will be taken from staff, experts who have used the system, that information will be collected and written into a report along with the direction of development Finally, we will hand over the project back to the climate center for use The aim of this project is to reduce the amount of data stored on a daily basis, use machine learning to predict the weather, increase accuracy, and help with real-time processing - A self-learning system - The system makes predictions on behalf of experts - Minimize the amount of input data Time Phase Plan Description Analyze goals, systems, technology to be used, feasibility Risk plan Learn similar systems Research data of current system Building final plan Building AI Run in actual system Evaluation and closing Get the request of the staff, experts at the center Analysis requirement Design system Implement (coding) Add to current system Testing Testing in system at center Get feedback of staff Evaluate Write report Development direction and hand over Communication Time 15/8 - 21/8/2022 day day day days day 21/8 - 19/9/2022 day day days 14 day day day 19/9 – 07/11/2022 Status Done 07/11/2022 – 14/11/2022 day days days day Awating Done Processing Our team will have face-to-face meetings and some online meetings with Google Meet Daily reports will be sent via chat group on Zalo Type Daily meeting Daily Report Analysis (Plan) Research Data Analysis Requirement Design Implement Testing Frequency or Purpose times 3-5 times/week Reporting and implementing new tasks Google Meet Every day times Zalo Offline Daily progress report Analyze problems to develop plans and risk plans times Working with staffs, experts to understand data times Segregate the task of taking user requests and analyzing them Every day (in Building system weeks) Every day (14 Test system days) Method Offline Offline Google Meet + Offline Google Meet Daily test system Get feedback Evaluate Handover 20-25 times/ Test system and check work progress weeks times Get feedback of user times times Offline Offline Evaluate and write report of project Offline Handing over the system to the climate Offline center leader Risks No Name Requirement Not enough people Project schedule Technology Cost Many errors Description Infulence Consequence level Collect or evaluate user High Depending on the requirements, wrong stage that affects the project purpose project differently, it can affect the whole project Temporary leave due to Low Slow down the illness or other causes project progress (more than people) Important in-person Low Slow down the meetings (possibly due to progress weather) The selected technology Medium Back to the planning is not suitable for the stage system Types of costs incurred Low Lack of budget such as hiring experts, buying equipment After actual testing, there were more errors than expected Medium Fix bugs that slow progress, waste valuable time and data Solution Work with many experts to understand, collect clear requirements, seriously analyze Find replacement personnel from the beginning of the project Prepare another temporary replacement plan to avoid delays Work with development experts to understand, choose the right technology Prepare a contingency budget at the beginning of the project (about 10-20% of the main budget) Detailed test planning Resources Tools: Upgrading system hardware (chips, heatsinks, etc.) Google Meet, Zalo Development tools for AI Human: persons: senior, junior AI, tester Cost estimation No Type Salary Detail senior (3000$/month), junior (1500$/month), tester (1000$/month), time is month Upgrade the system to be able to integrate AI technology Upgrading system hardware Hire an Hire experts to advise on technology, development direction expert Learn Working on projects using similar technology more technology Risks The cost to deal with the risks if they arise, it is about 10% of the total cost Total Planning WBS Cost 25500$ 25000$ 1000$ 500$ 5000$ 57000$ Figure 1: WBS This WBS is divided into phases: Planning, Building, Run the actual system, Evaluation and closing In the planning phase, the first thing is to plan and analyze the project The problems analyzed are the advantages and disadvantages of the old system, the solution for the disadvantages, the technology for the solution, the feasibility of the solution Next, find out the possible risks (natural and man-made), then create a risk plan including remedial methods Finally, the data of the current system is studied, as a premise to apply to the new system The building phase will learn the requirements of employees who use the current system, and then analyze those requirements to build functions for the new system Next is system design, building wireframes, diagrams, etc Based on those designs to build the system (programming) Then add to the existing system to test the new system The stage of running the actual system is simply observing and recording the results and performance of the AI when it comes to actual work The Evaluation and closing phase will take user reviews (mainly data) to get information about the accuracy and usefulness of the project Then write an evaluation of the results, write a report and future development directions based on the evaluated information Finally, handover to end the project Gantt chart Figure 2: Gantt chart The Gantt chart designed for the project is also divided into phases like WBS, the time span is about months (larger because there are days off) The first is the planning phase which will include project planning activities (1 day), risk plan (1 day), research data and real project – activities at the center and projects at other enterprises, so it will take days, then spend day to develop the final plan Phase building includes the work of taking requirements from users and analyzing them (2 days), system design (3 days because of a lot of work), Coding (2 weeks), Adding to the system and testing (2 days) day) Phase run the actual system will take about weeks to achieve accuracy as well as gather enough parameters for the system Phase Evaluation and closing includes taking user feedback, then evaluating and writing reports, building future development plans Finally, hand over to the climate center All wrapped up within week All phases will be performed by Team member (our team) Primary research Methods of building primary research Primary research includes interviews and surveys, interviews will be used qualitative research methods because it will take people's subjective information, information related to feelings, impressions and usually not use data For the survey, the quantitative research method will be used because it will present the data to the survey takers, take those data and analyze The qualitative research method used is In depth Interviewing, it is optimal for collecting data about personal views and experiences and especially understanding of the problem The quantitative research method used is survey, it is used to ask closed questions about the issues to be collected, the collected data will be data such as rate, level, etc Purpose of Primary research The purpose of primary research is to collect data from respondents, interview about current data problems, find out the influence and level of interest of the surveyors on the problem of data pollution Rely on their experience and understanding to have more contributions to the project The post-project survey will give the most accurate data for the project as well as the system, confirm the feasibility in practice and the level of user satisfaction Most importantly, survey takers will make assessments and suggestions for future development based on user roles Overview about Interview/Survey (Who, What, Why, When, Where) The purpose of the interview was to determine the feasibility of the project as well as the complexity, data-based, and user confidence in the project The audience of the interview are the staff, experts at the National Hydrometeorological Administration, they are the users of the system later, their requirements are the core requirements for the project The interview will be conducted in the 12th week of the project (the last week) to get the best evaluations List of interview question What's your name ? (Closed question) What is your position in the agency? (Closed question) Do you know anything about AI? (Opened question) Weather data is increasing, polluting the environment, what you think is the solution? (Opened question) Is the AI project to control the weather possible? (Closed question) After testing, you find this project useful? Does it predict correctly? (Closed question) What you want the system to be improved in the future? (Opened question) Summary about interview Answer: Most of the user answers not understand the problem of pollution by data However, they highly recommend this hybrid AI project Through the interview, users provided the necessary information about their understanding of AI, the problem of environmental pollution caused by the data, the feasibility of the system, and the usefulness of doing so actual work In addition, users also provide requirements for future development, thereby building a development plan List of survey question The survey includes personal information and some questions about the project We received 10 replies from users (due to the relatively small number of users) Form:https://docs.google.com/forms/d/e/1FAIpQLSd7WoEOoYn4GVRO9G7uUHms7dyW4fJ1mJZA dMPSJHVLay7YzA/viewform?usp=sf_link Figure 3: Suvery question (1) Most are system administrators (staff), in addition there are experts and interns (Closed question) Figure 4: Survey question (2) Basically the system has high accuracy in weather prediction (Closed question) Figure 5: Survey question (3) The project reduced the amount of input data stored by at least 70% compared to before, which is an impressive number (Closed question) Figure 6: Survey question (4) AI also helps reduce human resources to operate and manage the system (Closed question) Figure 7: Survey question (5) Ultimately improving in the future, the problem is speed in real time, accuracy (Opened question) Summary about survey Basically, based on user surveys, the AI project was able to successfully complete the real tasks of the system operators, accurately filtering the necessary data Things to improve in the future are speed, accuracy Evaluation Basically, the project has completed well and met the purposes set out in the plan In particular, it solves the problem of input data well, contributing to reducing environmental pollution caused by data In addition, it can also replace people, reducing the agency's resources This AI system will mature with time and daily data Secondary research In secondary research, we will explore available projects on data reduction, find out their advantages and disadvantages for the environment and people In addition, articles were also explored to clarify these issues 10 Purpose The main aim of the secondary research is to explore the scientific reports on AI technology for realtime systems as well as the difficulty and potential of AI for data management Find out the advantages and disadvantages of the system and give appropriate development direction in the future List of articles/ books [1] Jiang, F., Jiang, Y., Zhi, H., Dong, Y., Li, H., Ma, S., Wang, Y., Dong, Q., Shen, H and Wang, Y., 2017 Artificial intelligence in healthcare: past, present and future Stroke and vascular neurology, 2(4) [2] Alreshidi, E., 2019 Smart sustainable agriculture (SSA) solution underpinned by internet of things (IoT) and artificial intelligence (AI) arXiv preprint arXiv:1906.03106 [3] Musliner, D.J., Hendler, J.A., Agrawala, A.K., Durfee, E.H., Strosnider, J.K and Paul, C.J., 1995 The challenges of real-time AI Computer, 28(1), pp.58-66 Summarize Artificial intelligence in healthcare: past, present and future The use of AI in healthcare, presents various healthcare data that AI has analyzed, and surveys the main types of diseases for which AI has been deployed The two devices used are ML and NLP with their respective techniques Patient-related information will be collected and used by AI to make clinical diagnoses It is also sent to the doctors to get feedback With the advantage of rich data, AI can help doctors study diseases or focus on important data [1] According to the report, the biggest obstacle to data management AI is still legal regulations and certain risks when encountering rare diseases However in the future it will become a great tool to help hospitals [1] Smart Sustainable Agriculture (SSA) Solution Underpinned by Internet of Things (IoT) and Artificial Intelligence (AI) According to Alreshidi, this paper has established the importance of employing recent and advanced computing technologies in the agricultural sector, in particularly AI and IoT Agriculture is considered central to the survival of human beings Supporting the current practices of traditional agriculture with recent IoT/AI technologies can improve the performance, quality and volume of production This study has reviewed the existing IoT/AI technologies discussed within the main research journals in the area of agricultural Furthermore, it categorized the main domains of smart, sustainable agriculture, i.e human resources; crops; weather; soil; pests; fertilization; farming products; irrigation/water; livestock; machines; and fields The major contribution of this paper concerns the AI/IoT technical architecture for SSA, leading to an emphasis on the research and development of a unified AI/IoT platform for SSA, to positively resolve issues resulting from the fragmentary nature of the agricultural process Future work will include investigation of the process of implementing AI/IoT technologies for SSA by applying 11 the proposed AI/IoT technical architecture in the form of the prototype of a unified platform on real test cases This will identify the relevant strengthens and weaknesses for further improvement and enhancement [2] The major components of intelligent, sustainable agriculture, such as human resources, flower color, weather, soil, pests, fertilizer, agricultural goods, irrigation water, rising Castle, machinery, and fields, have also been categorised The deployment of AI/IoT technologies for SSA will be studied in the future, and actual test cases will be used to apply the suggested AI/IoT engineering architecture as a prototype of a unified platform [2] Basically, this article has analyzed the application of AI in agriculture including weather problems, it combines with IoT to process real-time data that is resource data , water, machinery, weather and make the best judgments It also doesn't store unnecessary data and doesn't pollute data, but it's still not perfect at the moment The challenges of real-time AI The author defined and discussed three key methods for creating real-time AI systems, building on earlier work The architectural structure of real-time and AI techniques, as well as the distinction between various approaches based on their performance goals, are all ways that this work improves upon the prior classical approach Real-time AI is still in its infancy, therefore picking the right system architecture for a given application necessitates a detailed explanation of its performance objectives and design aspects We anticipate that such descriptions will begin with our traditional approach to real-time AI We have highlighted numerous tough research topics crucial for the creation of the next generation of real-time AI by analyzing the functional restrictions these various designs put on its constituent parts [3] The demand for more potent and practical real-time system support technology is rising quickly as more sophisticated intelligent algorithms are applied to control systems in crucial sectors We described a number of system support areas in Part that will be essential for the quick creation of the next generation of intelligent real-time control systems Similarly, we discover that the broad adoption of AI methods for real-time domains will call for novel strategies that include many of the time-tested searchbased methods investigated in this area [3] Reactivity, incremental algorithms, balanced decision theory methods, and other novel approaches will be needed to intelligently adapt to the demands of these settings, as was covered in Section significant area Therefore, this article can help to some extent in future research focused at raising the standard of art Existing systems provide a solid early foundation, but they are unable to fully coordinate real-time response and intelligent, intentional behavior, as implied by real-time AI's full consciousness This places strict limitations on real-time or AI components, limiting their applicability to specific domains Some of these limitations are lifted by the real-time AI collaborative method, but most current 12 implementations are still master-slave systems where the subsystem is enslaved and only executes the plans supplied [3] Long-term, we anticipate that combining real-time research with AI will result in systems that are more integrated while preserving the same features for job planning, scheduling, and execution For activities that are mission-critical, assured performance is necessary Naturally, how exactly this is done will be a byproduct of subsequent effort However, it is obvious that the improvements in the areas we have mentioned will get us a step closer to achieving our overarching objective of creating a reliable but effective goal-directed system understand and accommodate its own environmental dynamics and resource limitations [3] Evaluation After the above studies, I have understood the causes of the increase in data and its harmful effects on the environment The huge amount of data increasing day by day can be completely reduced by AI solutions by applying it to the system to process input data at real-time speed, but it also needs a system The system is powerful enough to handle, so a large application system is most efficient If the project has a larger amount of time, we will explore more similar projects to find out the major advantages and disadvantages of other projects, thoroughly understand the user requirements for both current and future, from which to build the best system 13 Powered by TCPDF (www.tcpdf.org) Index of comments 2.1 Task 1: - Climate Management is presented - Main aim and objectives of the projects are reported more clearly in this revision - WBS and Gantt Chart are reported and explained in this revision Task 2: - Primary and secondary researches are reported more clearly in this revision Powered by TCPDF (www.tcpdf.org) ... and hand over Communication Time 15 /8 - 21/ 8 /2022 day day day days day 21/ 8 - 19 /9 /2022 day day days 14 day day day 19 /9 – 07 /11 /2022 Status Done 07 /11 /2022 – 14 /11 /2022 day days days day Awating... This AI system will mature with time and daily data Secondary research In secondary research, we will explore available projects on data reduction, find out their advantages and disadvantages... include project planning activities (1 day), risk plan (1 day), research data and real project – activities at the center and projects at other enterprises, so it will take days, then spend day to