Meta-heuristic method applied to a virtual networking platform

Abstract

The present invention relates to a computational logic process that uses a meta-heuristic method divided into two blocks of processes based on PSO and ACO, interconnected in order to analyse and define more precise combined results between a database of professionals, available jobs, professional preparation pathways, and gender inclusion.

Description

[0001] META-HEURISTIC METHOD APPLIED TO VIRTUAL NETWORKING PLATFORM FIELD OF THE INVENTION

[0001] The invention focuses on the field of computational methods that use meta-heuristic logic to segment collected data and segment the best "scenarios" in a logical decision-making process to return satisfactory results under a given computational process. Therefore, the respective computational logical process that uses a meta-heuristic method divided into two processing blocks with distinct but intertwined languages ​​to analyze and delimit more accurate combinatorial results among a database of professionals, available vacancies, and professional preparation paths. TECHNICAL BASIS

[0002] The construction industry plays a fundamental role in society, and women's participation in this sector is crucial for several aspects. In recent years, we have seen significant growth in female representation in the construction industry, and this change has positive impacts across several sectors.

[0003] First, the inclusion of women in the construction market contributes to a greater diversity of ideas, approaches, and perspectives. Gender diversity promotes innovation and creativity, as it brings different experiences and skills to the sector. This results in more comprehensive and effective solutions, meeting the diverse needs of society.

[0004] Furthermore, women's participation in the construction industry contributes to gender equality in the workplace. The construction industry used to be traditionally male-dominated, but this dynamic is gradually changing. Recognizing the value of women's work in the sector not only promotes equal opportunities but also reduces pay inequality and empowers women to pursue careers in a historically male-dominated field.

[0005] Another relevant aspect is the positive economic impact of women's presence in the construction industry. With a diverse workforce, companies can leverage the talent potential of both genders. This improves productivity and efficiency, leading to stronger economic growth in the sector, which, in turn, contributes to the development of the economy as a whole.

[0006] Furthermore, it serves as a role model and inspiration for future generations of women by demonstrating that women can be successful and welcome in all fields, including those considered non-traditional. This encourages more women to consider careers in construction, further expanding diversity in the sector and ensuring a more equal and inclusive future.

[0007] The inclusion of women in the construction market is not only a matter of gender equity, but also an opportunity to apply innovative approaches to human resource management, such as the use of Software as a Service (SaaS)-based recruitment and selection systems (PSO). This SaaS recruitment model can be customized to ensure equitable representation of women in the construction industry.

[0008] The (PSO) is a method that can be adapted to optimize talent searches while promoting gender diversity. By including variables that consider candidates' backgrounds and skills, as well as equal opportunity policies, the SaaS recruiting system can identify and prioritize candidates who bring a valuable skill set and a diverse perspective.

[0009] Furthermore, the (PSO) can be used to optimize the balance between gender equality goals and the operational needs of construction companies. The system can be configured to ensure that women are included in all stages of the recruitment process, from identifying qualified candidates to the interview and final selection.

[0010] Applying a metaheuristic model like the (PSO) to a SaaS recruitment system provides an efficient, data-driven approach to including women in the construction industry. This not only meets gender equality goals but also contributes to innovation and diversity of ideas in the sector, strengthening companies and driving the construction industry's progress toward a more equal and inclusive future.

[0011] Combining the inclusion of women with advanced recruitment methodologies not only strengthens the construction industry but also serves as an example for other sectors, demonstrating that gender diversity is a valuable asset that can be boosted through technology and innovation. This, in turn, creates a broader positive impact on society and the economy as a whole.

[0012] An important example of a method for optimizing job selection processes and professional profiles is the use of process automation using metaheuristic algorithms that utilize key variables, such as candidate skills and experience, specific job requirements, and evaluation criteria, to find the best matches. These algorithms leverage a variety of advanced algorithms, such as genetic algorithms, particle swarm optimization, and tabu search algorithms, to find optimal or satisfactory solutions in a complex search space.

[0013] These intelligent approaches enable companies to efficiently and effectively match candidate skills to job requirements.

[0014] Through data analysis and machine learning, metaheuristics identify patterns and relationships between variables to continuously improve the selection process. This results in more accurate matches and the identification of professionals who perfectly fit available positions, while also enabling candidates to find job opportunities that align with their qualifications and goals. This plays a crucial role in modernizing and improving job selection processes and professional profiles, making the job market more efficient and aligning companies' needs with candidates' skills more accurately and quickly.

[0015] Furthermore, variable-based metaheuristics are particularly valuable in a dynamic and ever-evolving job market. As company demands and employee skills evolve, these flexible approaches can adapt quickly, finding the best matches even in ever-changing scenarios.

[0016] These techniques can also take into account specific variables, such as cultural preferences, geographic location, and career objectives of candidates, as well as company-specific criteria, such as corporate values ​​and business goals. This makes the selection process more personalized and aligned with the individual needs of both parties.

[0017] As variable-based metaheuristics continue to evolve with technological advancements, we can expect the job market to become more efficient, inclusive, and accurate in matching job openings and job profiles. This benefits both companies, who can find highly qualified professionals, and candidates, who can discover opportunities that perfectly fit their career paths, making the selection process a more satisfying and successful experience for everyone involved.

[0018] Furthermore, variable-based metaheuristics also contribute to diversity and equality in the workforce. They can be configured to prioritize diversity criteria such as gender, race, and ethnicity, ensuring that companies are promoting inclusion and equity in their hiring processes.

[0019] These approaches are also valuable in addressing talent shortages in certain fields, allowing companies to identify candidates with transferable skills that can be tailored to meet specific job needs.

[0020] As companies recognize the potential of variable-based metaheuristics, they are increasingly investing in technology and partnering with data analytics and machine learning experts. This puts job selection and job profiles on a path of constant evolution, where the ideal match between candidates and employers becomes an accessible and accurate reality.

[0021] The multi-block, metaheuristic processing logic method represents a revolutionary transformation in the way companies and professionals find ideal matches in the job market. It offers efficiency, personalization, inclusion, and a precise alignment of skills and needs, contributing to a stronger and more dynamic job market. As these technologies continue to develop, we can expect a future of even more effective and efficient job and profile selection.

[0022] The process identified as "US7703071" is notable for its ability to build an initial analysis model, which comprises IBM WebSphere Business (WB) notation, based on at least one legacy business process model. This initial analysis model is essentially a business process model that represents the subdivision of an initial legacy business process model.

[0023] Within this process model, event-driven process chain notations are incorporated, as well as a variety of tasks that group components of the initial WB model, which in turn is derived from the first legacy business process model. These tasks are logically interrelated and involve performing a componentization of the initial WB model.

[0024] The method involves making adjustments to the control flows and process components of the initial WB model and modifying the decision conditions based on data in the second analysis model, resulting in the definition of a design object model.

[0025] The process, referred to as "US20200050983," describes a system and method combination for generating a business workflow process. This is accomplished by recording a series of real-time screenshots that capture the steps required to complete a business workflow process in one or more business applications. The recorded information is stored in a workflow data file. These screenshots include the capture of technical attributes during the recording process, such as control identifiers, control types, control names, class names, control data values, paths, process names, process IDs, process descriptions, control application names, and control application screen names.

[0026] The "US11373224" lawsuit refers to a producer interface, which can be an application, service, or Application Programming Interface (API). This interface consists of a set of modules that, when accessed, execute functions or events related to business concepts relevant to a business entity, such as goods, services, users, accounts, and bidding transactions. Directives or actions are generated by resources, which can be automated programs, services, or users.

[0027] In a specific modality, statistical information about the events generated, their distribution and processing can be gathered in generic objects or formats by each of the application's interfaces.

[0028] This statistical information can be viewed in real time on screens with configurable formats or stored in data logs for subsequent evaluation and reporting.

[0029] Related virtual models are primarily composed of preconfigured items, such as colors and item models for virtual assembly. The final result of the visual virtualization of the model aims to relate the possible combinations predefined by the system. Therefore, they bear no similarity to the present innovation. BRIEF DESCRIPTION OF THE FIGURES

[0030] The innovation of the subject of the patent application can be better understood through the following figures: ✔ Figure 1 shows the steps of the virtual platform processes referring to block 1 of the algorithm; ✔ Figure 2 shows the steps of the virtual platform processes referring to block 2 of the algorithm; ✔ Figure 3 shows the flow of the algorithm (block 1); ✔ Figure 4 shows the flow of the algorithm (block 2); ✔ Figure 5 shows the diagram of the complete algorithm; ✔ Figure 6 shows the flow of the complete algorithm (block 1) + (block 2); ✔ Figure 7 shows the continuous cycle of the management method applying block 1 and block 2 of the algorithm. GENERAL DESCRIPTION OF THE INVENTION

[0031] This innovation comprises a computational method (algorithm) embedded in a computational platform using decision heuristics to prospect projects in the construction sector, connecting companies, stakeholders, and professionals to promote the inclusion of women in construction projects. The platform uses advanced algorithms to analyze both the technical and human skills of women interested in working in the construction market. It considers past experience and technical certifications, and also assesses characteristics such as teamwork, communication, and problem-solving. By efficiently combining these criteria, the platform helps identify ideal profiles, promoting more accurate and inclusive selection for the sector.

[0032] Furthermore, it presents an algorithmic method that returns as a final result the implementation of a networking platform for the construction segment aimed at women based on meta-heuristics (PSO) and (ACO) taking into account a series of variables to ensure efficiency and equal opportunities.

[0033] The major innovation behind the algorithm, in addition to the meta-heuristic logic incorporated, is the real-time monitoring of the participatory management of all stakeholders through an interface that demonstrates all the parameters of the selection process, allowing monitoring by all those involved in the project.

[0034] The integration of the meta-heuristic concept of Particle Swarm Optimization (PSO) with Ant Colonies (ACO) in a networking platform for gender inclusion in the construction industry represents an innovative approach providing synergy between algorithms. The platform becomes a dynamic ecosystem, where candidates and stakeholders collaborate efficiently to promote diversity in the construction industry.

[0035] On the one hand, the language-based algorithm (PSO), inspired by the behavior of migratory birds, guides candidates in their search for opportunities, optimizing their choices based on past experiences and specific goals. Each "particle" represents a professional seeking career growth. The coordinated movement of these particles reflects interactions on the platform, dynamically adjusting to changing individual needs and goals.

[0036] At the same time, the (ACO) concept works to strengthen collaboration and communication in the community.

[0037] To use an analogy, ants represent female candidates, and pheromone trails symbolize interactions and choices made on the platform. Ant colony-based algorithms (ACO) determine activity selection, dynamically adapting to participants' needs and tendencies.

[0038] The ants, or candidates, follow the trail of available possibilities, influenced by binary weights that represent the quality of the interaction. The pheromone's evaporation rate controls the persistence of opportunities. This ensures that the platform learns and adapts over time.

[0039] Influence weighting emphasizes the importance of past experiences in activity selection, while pheromone weightings highlight the influence of historical events on participant decisions. These adaptive parameters ensure the platform's continuous adaptation to reflect the community's changing needs and priorities.

[0040] The pheromone deposit rate represents the positive impact each woman can have on the community by storing valuable information and building stronger pathways for future participants. In this way, the platform becomes a learning and development resource where members share their experiences and highlight relevant opportunities for other women. This exchange of information provides the basis for generating valuable data that will aid in the integration and opportunities of women.

[0041] The 2 blocks of language-based algorithms (PSO) and (ACO) intertwine, using binary weights and evaporation rates to adjust the importance of past interactions in decision making.

[0042] The platform not only connects women with mentoring, training, and project opportunities, but also offers a holistic approach to gender inclusion in the construction industry. Awareness-raising events, hands-on workshops, and the highlighting of success stories strengthen the inclusive culture.

[0043] Ultimately, the application of the ant colony algorithm (ACO) to this networking platform not only catalyzes gender inclusion in the construction industry but also sets a groundbreaking standard for promoting diversity and equal opportunities in historically male-dominated industries. DETAILED DESCRIPTION OF THE INVENTION

[0044] This innovation comprises a computational method (algorithm) inserted into a computational platform using two blocks of processes that operate simultaneously, the first of which operates in the execution of the stages of ongoing projects, while the second operates in the selection of candidate profiles, most suitable for each vacancy available in the ongoing project.

[0045] Figures 1 and 2 demonstrate the flow of activities developed in a project, combining the performance of blocks 1 and 2 of the algorithm; Figure 3 presents the simplified flow of block 1 of the computational algorithm applied to the computational platform; and; Figure 4 presents the simplified flow of block 2 of the computational algorithm applied to the computational platform; Figure 5 represents the general structure of the algorithm; and Figure 6 represents, in a simplified way, the connection of the complexity and processing of the operation method of block 1 and block 2.

[0046] To execute an iteration of Block 1, we have: ✓ For each iteration i and j: 1. Choosing the Next Step: 2. “Ant” movement: ^^ = ^^^^^^^^^ℎ^^^^ ^^^^ó^^^^^^^^ ^^^^^^^^^ ^^^^^^^^ ^^^^^^^ ^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^3. Pheromone Update: ^^^^^^ = (1 − ^^) ∙ ^^^^^^ + ^^ ∙ ^^4. Global Pheromone Update: Breakdown for Specific Steps: ^^^^^^ = ^^^^^^^^^^^^çã^^ ^^^^^^^^^^^í^^^^^^^^ ^^^^^^^^ ^^^^^^^^ ^^^^^^^^^• The process continues until a defined number of iterations m is reached. • The values ​​of the parameters (α, β, ρ, Q, etc.) should be adjusted to optimize the efficiency of the algorithm.

[0047] For the execution flow of Block 1, the algorithm performs the following steps: ✓ Parameters: ^^ − ^^ú^^^^^^^ ^^^^ ^^^^^^^^^^^^^^^^ (^^^^^^^^^^^^)^^ − ^^ú^^^^^^^^ ^^^^ ^^^^^^^^^^çõ^^^^^ − ^^^^^^^^ ^^^^ ^^^^^^^^^^^ê^^^^^^^^^^ − ^^^^^^^^ ^^^^^^^^ ^^^^^^^^^^^^ô^^^^^^^ − ^^^^^^^^ ^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^çã^^ ^^^^ ^^^^^^^^^^ô^^^^^^^ − ^^^^^^^^ ^^^^ ^^^^^^^ó^^^^^^^^ ^^^ ^^^^^^^^^^^^ô^^^^^^^^^^^ − ^^^^^^^^^^^^^^^^^^^ ^^^^ ^^^^^^^^^^^^^^ ^^^^ ^^^^^^^^^^^^^ ^^^^ ^^^^^^^^^^^^ ^^^^^^^^^^^ ^^^^^^^^^^ ^^✓ Initialization: • Initialize pheromones in all steps with an initial value (^^0); • Each "ant" (project step) starts at the "Meeting with Leadership" step. ✓ For each iteration i from 1 to m, represented in the following sequence: 1. Choosing the Next Step: 2. “Ant” movement: ^^ = ^^^^^^^^^ℎ^^^^ ^^^^ó^^^^^^^^ ^^^^^^^^^ ^^^^^^^^ ^^^^^^^ ^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^3. Pheromone Update: ^^^^^^ = (1 − ^^) ∙ ^^^^^^ + ^^ ∙ ^^4. Global Pheromone Update: 5. Breakdown for Specific Steps: ^^^^^^ = ^^^^^^^^^^^^çã^^ ^^^^^^^^^^^í^^^^^^^ ^^^^^^^^ ^^^^^^^^ ^^^^^^^^^^^Breakdown for Specific Steps: • Leadership Meeting (Weeks 1-2): ^^^^^^ = ^^^^^^^^^^^çã^^ ^^^ ^^^^^^^^^^^â^^^^^^^^^ ^^^ ^^^^^^^^^^^^ ^^^^^^^^^^^ ^^ ^^^^^^^^^^^^^ç^^• Identify Responsible Team (Weeks 1-2): ^^^^^^ = ^^^^^^^^^^^çã^^ ^^^ ^^^^^^^^^^^ê^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^çã^^ ^^^ ^^^^^^^^^^^^^^^^^• Set Specific Goals (Weeks 1-2): ^^^^^^ = ^^^^^^^^^^^^â^^^^^^^^ ^^^ ^^^^^^^^^^^^çã^^ ^^^ ^^^^^^^^^^^ ^^^^^^^^^^í^^^^^^^^• Internal Diagnostic Assessment (Week 3-4): ^^^^^^ = ^^^^^^^^^^^çã^^ ^^^ ^^^^^^^^^^^^^^^^^^ ^^^^ ^^^^^^^^^^^^^ ^^^^^^^^^^^ ^^^^^^^^^^^ ^^^^^^^^^^^ ^^^^^^^^^^^ ^^^^^^^^^^^^ •Internal Lecture (Week 3-4): ^^^^^^ = ^^^^^^^^^^^^çã^^ ^^^ ^^^^^^^^^^^^^^ ^^^^ ^^^^^^^^^^^^^^^ ^^^^ ^^^^^^^^^^^^^ ^^^^ ^^^^ ^^^^ ^^^^ ^^ê^^^^^^^• Selection of Beneficiary Entity (Week 3-4): ^^^^^^ = ^^^^^^^^^^^^^çã^^ ^^^ ^^^^^^^^^^^^^^^ ^^^^^^^^^^^^ ^^^^ ^^^^^^^^^^^^^^^ ^^^^^^^^^^ℎ^^^^^^• Present Interest Mapping (Week 1-2 of Month 2): ^^^^^^ = ^^^^^^^^^^^^çã^^ ^^^ ^^^^^^^^^^^ê^^^^^^^ ^^^ ^^^^^^^^^^^^^^^^^^^çã^^ ^^^ ^^^^^^^^^^^^^^^^^^^^• Assemble Stakeholder Network (Week 1-2 of Month 2): ^^^^^^ = ^^^^^^^^^^^^^çã^^ ^^^ ^^^^^^^^^^^^^ê^^^^^^^^ ^^^^^^^^^^^çã^^ ^^^ ^^^^^^^^ ^^^^ ^^^^^^^^^^ℎ^^^^^^^^^^^^ •Budget Planning (Week 1-2 of Month 2): ^^^^^^ = ^^^^^^^^^^^çã^^ ^^^ ^^^^^^^^^^^ê^^^^^^^ ^^^ ^^^^^^^^^^^^^^^^^^^^^ ^^^^ç^^^^^^^^^^á^^^^^• Official Project Launch (Week 3-4 of Month 2): ^^^^^^ = ^^^^^^^^^^^^çã^^ ^ ...^• The process continues until m iterations are reached. • The parameter values ​​(α, β, ρ, Q, etc.) must be adjusted to optimize the efficiency of the algorithm.

[0048] To execute the Block 2 process, the algorithm uses the following variables to determine the search funnel, data grouping, and determine optimal solutions: • Variables ✓ Experience in civil construction: Experience in construction projects, considering variables such as years of experience and types of previous projects; ✓ Technical skills: Specific skills related to expertise such as masonry, carpentry, plumbing, electrical, welding, etc.; ✓ Academic background: Technical knowledge intrinsic to the candidates' training, such as diplomas, certificates, and training courses in civil construction; ✓ Geographic location: Proximity of candidates to workplaces to optimize logistics and reduce travel time.• Restrictions ✓ Schedule availability: Candidates' availability for different work shifts, including variables such as daytime, nighttime, weekends, etc.; ✓ Interest in Heavy Construction: Level of interest and predisposition to specific heavy construction vacancies.

[0049] By applying the variables and restrictions in a modality of the invention, the algorithm will present the optimal configuration for executing the project, strictly observing the restrictions imposed by the variables and opting for the scenario that best harmonizes all the variables present in the logic developed for the user, simultaneously ensuring the processes described below:

[0050] Step 1: Candidate Data Table 1. Candidate Data

[0051] Step 2: Objective Function and Coefficient ✓ The objective function we want to maximize is a weighted combination of these features: ^^(^^) = ^^1 . ^^1 + ^^2 . ^^2 + ^^3. ^^3 + ^^4. ^^4✓ The weighting coefficients are values ​​associated with each variable in the objective function, in this case, for the features: ^^1 = ^^2 = ^^3 + ^^4 = 1

[0052] Step 3: Restrictions • Availability (^^1(^^)): ≤ 0 ✓ Candidate 1: Available for any schedule. ✓ Candidate 2: Available for daytime schedules only. ✓ Candidate 3: Available for evening and weekend schedules. ✓ Candidate 4: Available for daytime schedules. • Interest in Heavy Construction (^^2 ( ^^ ) ): ≤ 0 ✓ Candidate 1: High interest (value 1). ✓ Candidate 2: Medium interest (value 0.5). ✓ Candidate 3: Low interest (value 0.2). ✓ Candidate 4: High interest.

[0053] Step 4: Algorithm (PSO) Table 2. Parameters (PSO)

[0054] Step 5: Particle Update (PSO)

[0055] Step 6: Stopping Criteria

[0056] Step 7: Final Selection and Interpretation

[0057] The invention modality represents a possible optimal solution found by the algorithm, indicating the final selection of a candidate with 6 years of experience, a technical skill level of 0.88, a location within 8 km of the workplace, and a bachelor's degree. This interpretation helps understand the characteristics of the team selected based on the established criteria.

[0058] To connect the mathematical model of block 1 (ACO) with block 2 (PSO), we can consider that the result of the steps in block 1 will influence the search for opportunities and projects in block 2. Below is the representation of the algorithm for the integration of Block 1 (ACO) with Block 2 (PSO): •Initialization of (PSO): 1. Initialize the number of particles P to represent different design configurations. 2. Each particle of (PSO) will have a dimension D that represents the different design stages. • Representation of Particle of (PSO): 1. Each particle of (PSO) will have a vector representation, where each component of the vector corresponds to the choice of a specific design stage.

[0059] Each will be a binary value indicating whether stage j of the project was selected (1) or not (0). • Update of the Velocity and Position of the (PSO): 1. The update of the velocity and position in the (PSO) will be influenced by the results of the ACO. The evaluation of the objective function of the (PSO) will take into account the performance of the design configurations found by the ACO.

[0060] ^^^^^^^^^^^^^^^^^^^ represents the influence of the (ACO) on the updates of the (PSO) particles. This term can be a function that considers the quality of the solutions found by the (ACO). • Evaluation of the (PSO) Objective Function: 1. The (PSO) objective function will be evaluated based on the quality of the design configurations represented by the particles. This will include a consideration of the influence of the (ACO):

[0061] The function ^^^^^^^^^^^^^çã^^^^^^^^^^^^^^^^(^^ ^^ ) evaluates the quality of the design configuration represented by the particle ^^ ^^ , and ^^^^^^^^^^^^^^^^ê^^^^^^^^ ( ^^ ^^ ) represents the influence of (ACO) on the algorithm.

[0062] The global best configuration ^^^^,^^^^^^^^from Block 1 is used as input to Block 2 (PSO).

[0063] Therefore: ✓ The search space of Block 2 is defined by the best stage configurations found in Block 1; ✓ The optimization of Block 2 seeks to find the best professional configurations for each stage of the project, considering the best global configuration of Block 1.

[0064] The interaction between the blocks occurs in the transfer of information about the best configurations of project stages, allowing Block 2 to refine the selection of professionals according to the optimized characteristics of Block 1.

Claims

1 / 1 CLAIMS 1) META-HEURISTIC METHOD APPLIED TO VIRTUAL NETWORKING PLATFORM revealing a complex computational meta-heuristic method characterized by a computational algorithm with decision logic applied to an online virtual platform to select qualified professionals according to the decision variables, as per FIGS. 1 and 2, to connect professionals (women) with a network of companies, mentors and institutions; 2) META-HEURISTIC METHOD APPLIED TO VIRTUAL NETWORKING PLATFORM, according to claim 1, characterized by the fact that the method is divided into 2 process blocks, with Block 1 being based on meta-heuristic logic (ACO), as per FIG. 3, and Block 2 based on meta-heuristic logic (PSO), as shown in FIG. 4, creating an interweaving between the 2 blocks, as shown in FIG. 5 and 6, to connect the best vacancies and professionals in the projects developed, as shown in FIG. 1 and 2.

Images