Badges Earned from Jan 2023 to Apr 2023 Events from Construction Software Related. — Part I
Construction Software — Risks, Regulation, Science & Technology, Tech Exhibition.
Provenance of the Startup, HEAD SHIP LTD.
During the period of 1998, my Dad, Mr. Vijayakumar Pookat, used to work for a couple of Construction Companies that involved Site Work in Dubai, Sharjah, Abu Dhabi and Al-Ain, the Emirates of UAE.
The same period when a Red Maruti Esteem was driven by my Grandfather, Shri. C. Shivashankara. Menon, through his employment at Enarc Construction, a book was bought for Training himself during the job-shift from Small Buildings and Quarters Construction at Neyveli Lignite Corporation (NLC).
The book was on Piling and Construction Foundations, helpful for the Construction Foundation Sites.
I am a Technical Expert in Database Development and Programming. Even-though I was unfortunate to not use a Dictionary in my Game designed by me and inspired from an ITV Programme, Brainteaser, also Word-Pyramid.
The importance of a Dictionary in terms of a Data Structures or within GoTo Statements of a Borland C++ Programme created in a Floppy Disk, was known after speaking to My Computer Science Teacher about the usage of hard-coded patterns.
The hard-coded patterns matched two-three possibilities of words which would appear alternatively after giving a Hint down the Pyramid after the previous word-occurence.
A Construction Foundation Dataset has been started to collect from 1998 onwards and it involves many Construction Site Images, from sites of Employments he has had. Aggregated set of images involve at least 14k+ Construction Site Images from 2000 onwards in the Dataset. These images alongside other Internet Images collected manually are used to develop:
The Idea KlinterAI
1. Drift Algorithms
KlinterAI is an idea that is developed from a perspective of a Drift-Detection and Drift-Adaptation Project done during MSc Dissertation. The idea evolved from having a belief that through Mathematical Representations, Mathematical Formulations and Models one could achieve the Drift to be Detected to a good amount of Accuracy.
2. Naming of the Logo
KlinterAI was accepted as a name of the Brand representing the company for a Construction Software Product from a name “Clinker” — which is also the Residue of Construction Concrete Manufacturing formed in the form of globules.
3. Integrating Observability
KlinterAI is also shaped from the Observability Theme popular from 2022 onwards. The sources state Observability is considered as an Element of SecOps, AIOps and ITOps. Observability is compared against Monitoring to have a higher Uptime of 99.99%.
Replicability — Network & Software — 99.99% Uptime.
4. Backup Transfer of Enterprise Data over Network
During my Final Internship at VINCI plc, I worked with Jim FitzPatrick who was the IT Lead working with Inventory as well as RiverBed SteelHead for Network Transfer over Remote Sites. The sites connecting Southern-England and Northern-side of Offices are installed with Network Infrastructure copying Employee Data as well as Corporate Data.
In this set of Exercises, I had came up with a Porters Value Chain representing The Backup Strategy.
In a certain Use-Case, the Backed-Up Data consisting of Sniffed or Surveillanced Data is Analysed to provide insights in the form of Business Intelligence and Reporting.
In this set of Exercises, I had came up with a Data Integration Assessment and Architecture.
The Catalogue System (which is also like a Repository) exposes to an ETL Tool.
5. Received an Admit into MRes. GeoEngineering
I received an admit into MRes. GeoEngineering to study about Soil, Ground, Terrains and other Topics relevant to the Ground-related Study and predicting the Age of Ground Water. What I learned during the period was that the place Central Valley of California had a subsidence region that is prevented by filling in Water through Ground-Water Recharge.
Some of the ideas such as Crack Detection and Quarry Activity Detection could be through the KlinterAI idea.
6. A Multi-Factor Productivity in the Construction Sector
In 2019, I worked on a Data Envelopment Analysis (DEA) Problem involving a Sleep-Quiz, that involved developing a scale for a set of parameters.
In this article, Biases in Artificial Intelligence (AI) — Fairness and Bias, I have expressed the DEA Algorithm as a combinator of Privilege and Service.
Such Privilege and Service is taken to be originated from Contracts of the Council, Governments and other Bodies.
- The term “Privilege” usually refers to the Consumer choice and how they see a Product or Service they buy from the Market,
- Whereas the term “Service” originates from Applying accepted Factors adjoining the Privilege Market embedding the Service Market through which the Vendors and Shop-Keepers take these into consideration.
I, believe to study the Buying Behaviour one needs to Understand how the Privilege and Service Markets are behaving over a Period of Consideration of Trade Contracts and Purchase Orders.
Regarding the Productivity Outputs from the Market-Sector output and Construction output, the below Productivity graph provides a steadily increasing Market-Sector output but a diminishing Construction-Sector output which is Multi-Factor with occasional hikes.
7. Movement of Material involving Material Purchase Orders and a Material with Labour Involved.
In one of the Images from the Construction Foundation Dataset, the Piling-Cage is being moved from one location to the other location using 2 Load-Carrying Cranes. This photograph is relevant to KlinterAI discussion because the idea of managing these Purchase Orders and its entry into the Site is discussed in its Business Idea.
A Later Safety Incident Monitoring and a Risk Register is implemented through Computer-Vision System of ML Models and Depth Cameras Using EquipAny.
8. Snag-List used in Computer Science through AirFlow and AppSheet helps List all Features and Requirements of a Software.
The Punch-Lists used in the Construction Sector: to record Safety Incidents are similarly observed in the Snag-List for listing all the Requirements.
Snag-Lists are used in AirFlow and AppSheet like Applications.
Through the Listing of Punch List, Computer-Vision could be used to detect any Safety Incidents and Minimize Risk at Site. Moreover, Daily Reports are assisted through the Idea, KlinterAI, which a Foreman in his work uses to interview the Workers under a Labour Contract.
9. The Dimensionality of Building Information Modeling (BIM) — 4D, 5D, 6D ad 7D.
The Building Information Modeling (BIM) is the main source repository for information and storage of Assets-related Information or Project-related Information.
In 4-Dimensional BIM for Construction Projects, Scheduling is important with a Stage-wise Development of Sub-Structures and Super-Structures simulated within BIM. As discussed over a webinar, this is orchestrated into an animation.
BIM Dimensions and Domains of Interest Shown Below:
1. — BIM 4D: Scheduling.
2. — BIM 5D: Costing.
3. — BIM 6D: Environmental.
4. — BIM 7D: Documents.
10. Circulation and Building Study & NEPAL DISASTER.
In one of the Projects conducted in Hyderabad, India a Hotel Building required Design & Environmental Analysis. For this, the Material Properties, the Selection of Glass as Building Materials and Construction was carefully made. It involved Visual Comfort Studies with a study on daylight sensitivity analysis with the Earth’s atmosphere.
Circulation and Building Study.
Based on a Paper describing about, Heating contained inside the Buildings, Circulation is affected by Walls, Internal Walls and Surface Area Exposed to the People inside. External Temperature will influence the choice of the Building Materials used.
BIM 6D as an Environmental Domain benefits Projects that records Carbon footprint. I had an opportunity to attend one of the meetings, with a company that promotes taking credits of CO2 emissions generated within a Construction, Railway or other Projects. The Credits are offered through Project Repositories that connect across an Enterprise.
Nepal Earthquake Disaster Study.
The Nepal Earthquake Disaster Dataset on drivendata.com was produced as an Exercise in a Module with the Applied Data Programming (ADP). ADP addressed several learning outcomes for our MSc Data Science course.
We had a good Data Science, Data Visualization and an Exploratory Data Analysis (EDA) Final Assignment with our Computer Science Faculty.
Azimuthal and Altitude Angles Study.
In one of my exercises, I posed a thought experiment to understand how Azimuthal and Altitude Angles affect Natural Lighting provided to the Buildings. Same time, there were discussion on putting a Skylight at the roof of the House, my Dad was Constructing.
There could be an influence on where to Install Solar Cells for Enabling Heating and Supply of Electricity.
11. DWG & DXF Formats — Floor Plan and Font-Design.
In 2010, I was involved with the Drafting of Floor Plan and designs of Windows, Chairs and Layout with the Sunbury CitizensAdvice.
My sister, Ms. Aswathy Vijayakumar, who did BArch. from a Malappuram University, MES Kuttippuram, was also involved in this Floor Plan Drafting with me after her MSc Urban Design at the University of College, London.
When I was also assisting the CitizensAdvice, Staines Branch with Travel Expenses Management and IT Support with Moving the Machines, Installing a Software, I came to know about Intuit Software for Accounting through Quickbooks.
Having a familiarity with AutoCAD Software, I started a New Project beginning to be online-based through a Web-Interface, using Node.js, Web-Canvas, Backbone.js with an idea of projecting that towards: the Node-Canvas Projects with a Cairo Backend.
The Convert-Drawing Application did some Prototype Drawings and was Published on a Project termed Enscalo.
DWG and DXF Formats were candidates for exporting the Final Design from the Front-end for interchangeability.
The Project Enscalo spoke about 5 Spatio-Temporal Properties of Aesthetics for a Residential Building:
1. Orientation
2. Circulation
3. Landscaping
4. Building Materials
5. AestheticsIn one of my discussions, with my Grand-Dad, Shri. C. Shivashankara. Menon, he expressed his interests in the Vedic City, Astronomical in nature and about 5 essential points in Architecture, as presented in the earlier paragraph.
Out of these realms, the project considers 3 factors involved in Building Design:
1. Light
2. Circulation
3. RetentionOne of my friends, Ms. Sunitha Menon, who is also an Architect has discussed with me about achieving a Desired Design Process (DESIRED) through a Project Methodology.
Out of these learning outcomes, I’ve segregated 4 typical Types of Design solving the Purpose of Defining the Layouts:
1. Isolated Layout
2. Open-Plan Layout
3. Designed Layout
4. Partial Layout12. QSR and OpenVINO.
The first time I am reading about QSRs (Qualitative Spatial Relations) is when I saw the Library — QSRLib. There are several Relations defined such as: RCC (Region Connection Calculus), AIA (Allen’s Interval Algebra), TPCC (Ternary Point Connection Calculus) and others.
They are defined by ArcTan Relations, Cosine and Sine Relations and also using Displacement Values.
The Leeds University Paper, also is related to the Paper I read: “Combining Spatial and Temporal Logics: Expressiveness vs. Complexity”
Qualitative Spatial Relations (QSRs) are used to understand pose changes, and detect any stencil operations. In our Team’s SAS Hackathon 2023, representing EquipAny, the computer-vision coding construct Stencil was introduced to the Team. Stencil was also dealt with earlier in one of My Webinars about Performance Matters at the Edge.
I introduced a Google Group, at klinterai.com, associated to my Company.
Observability by Parts.
Introducing some Math concerning Drift
Drift Detection is associated to Algorithms that work with: Video-based Drift Detection, Image-based Drift Detection, Signal-based Drift and Drift in Tabular Data Training/Inference.
- Product of Differentials — Information Theoretic.
First set of Representation of a Calculus Equation developed is for — The HyperParameter Drift due to Aging. Upon changing the Age of a given face, we’re able to detect the change in HyperParameter Tuning Parameter that is irrespective of Age and one with Age-wise Trained Model. This is presented in 2021–2022.
2. Interval Analysis
Interval Analysis shown here is produced within an Article in Medium, Simulation Optimization — Operational Methods, which involves analysis of Small Deviations through a Normalization Process to give rise to a Residual Value. A Residual Value is sometimes represented as the Condition Number, that takes an N-norm at the numerator and then multiplies by the inverse of the matrix. Such changes of a Model over a Large Interval Value could be analyzed by the Taylor Series.
3. Complex Analysis with Fourier Transform
Complex Analysis with Fourier Transforms shown here are two components — (1) For the Real Part, and (2) For the Imag. Part.
Such equations are derived from a set of equations as per the Article presented here. Individual Fragments of Data provide a value mapping towards the Entire Fourier Transform that is Digitally Represented.
4. Factor Analysis
In one experiment, while Studying the Machine Learning Topics through Udacity, from Stanford University ML Topics, I learned Programming the Factor Analysis (FA), Work with Math of Independent Components Analysis (ICA), work through the Math of SVM Analysis with Geometric and Functional Margins and about Expectation-Maximization (EM).
Through some Data generated through Light Intensity Available within a Building and its Rooms, I conducted Factor Analysis (FA).
5. Diffusion as a Dispersion Relation
Using the Momentum Conservation
I took the Napiers Momentum Conservation for the Mechanics of Fluid Flow as an example to understand if I can convert from “Sums” to “Products”. And eventually I was able to in the form of a Differential Equation (DE). This was achieved over a derivation in 2018 when I was learning Re-Inforcement Learning. The manner with which: such a derivation was performed was from a Random Walk with a probability of 1/4 to each direction.
By considering Diffusion, this was reduced to the size of an aperture or walls through which such a flow occurs. For Diffusion there is a rate with which the phenomenon occurs and describes certain equations that govern the flow.
Using Enthalpy
In the beginning of the Derivation, a Mass Flux Parameter is established to conclude a pluggable entity inside the Derivation. The entire setting is also defined in a Very Hypothetical Condition that Entropy and Specific Volume form part of the Same Partial Differential Equation giving Rise to the Rate of Change of Enthalpy.
6. Qualitative Spatial Relations (QSRs)
In Qualitative Spatial Relations, we look for Overlaps of Entities, Objects and Things in consideration. In a time-series dataset, Algebra such as the Allen’s Interval Algebra (AIA) takes shape due to interaction of two or more signals. I, once analyzed signal data using MCTS (Monte Carlo Tree Search) into a hierarchical result. I performed this using a Policy Optimization Technique inherited from Multi-Arm Bandits Technique, a part of the Re-Inforcement Learning Algorithms.
The Datasets I took were Speech Data Signal, where there is Randomness and OCR (Optical Character Recognition) Data. This was represented in a GitHub repository in 2020.
7. Sensitivity detected in Neural Networks
To plot one section of the behaviour of the Neural Network or Machine Learning Model trained, error was introduced into the image which resulted in a greater PSNR error and thereby the Knowledge Loss was plotted too.
To refer to this a Sensitivity Equation, explains how it varies with the introduction of an error to the Input Data.
In the Face Drift Algorithm, the PSNR error was considered the Knowledge Loss and the MSE (Mean Squared Error) was the Drift Severity.
An Insight into the Face Drift Algorithm and Explaining its Drift using Graphical Methods.
Overall Picture of the Math behind Drift and other related Processes
