System for multi-stage planning of construction processes and resource allocation

Abstract

A system for multi-stage planning of construction processes and resource allocation, consisting of: a central planning engine configured to receive input data, including architectural design models, structural constraints, procurement schedules, and historical performance indicators; a task decomposition processor that is operationally connected to the central planning engine and configured to generate a hierarchical construction task graph by decomposing macro-level construction milestones into mid-level and micro-level subtasks, with each subtask having time estimates, location identifiers, resource requirements, and mutual dependencies; a hybrid planning processing unit configured to resolve time and resource constraints across the entire task diagram; a resource coordination controller that is operationally connected to the central planning engine, wherein the resource coordination controller includes a real-time database of work units, machines and material stocks, each resource being tagged with attributes such as availability, usage history, operating status and spatial location; a multitude of distributed execution units distributed across the construction zones, each distributed execution unit comprising an embedded controller, sensor interfaces, task status processing logic, and communication circuitry, each distributed execution unit being configured to receive planning instructions from the central planning machine, execute localized control logic for task confirmation and resource activation, and transmit task execution data back to the central planning machine; an adaptive conflict resolution processing unit that is operationally connected to the central planning engine and configured to detect conflicts in task execution or resource conflicts, simulate alternative task-resource allocation scenarios using a real-time multi-agent model, and autonomously update the task graph with revised task sequences and resource allocations; and A dashboard for the construction process, configured to visualize task progress, deviations from the planned schedule, and resource efficiency metrics, with the dashboard also being able to receive manual override inputs or approve automated conflict resolution proposals generated by the adaptive conflict resolution module.

Description

Field of invention

[0001] The present invention relates to construction project management and, in particular, to a computer-aided system and a machine-integrated architecture for multi-stage planning and dynamic resource allocation in construction processes. The invention utilizes real-time data acquisition, predictive planning techniques, and the hierarchical coordination of tasks at various levels of construction activity—from site management to macro-level dependencies—thus ensuring the optimal use of labor, materials, machinery, and time in complex construction projects. Background of the invention

[0002] Construction projects are inherently complex, often involving multiple layers of interdependent tasks, diverse resource pools, and dynamic changes driven by environmental, logistical, and human variables. Traditional project management tools offer linear, Gantt chart-based tracking that fails to dynamically adapt to site conditions or allocate resources with granular, priority-based precision. Furthermore, conflicts between concurrent tasks in shared spaces (e.g., scaffolding vs. facade installation), material delivery delays, and the misallocation of labor and equipment can significantly impact schedules and budgets. Existing systems also lack integration between architectural design data (e.g., BIM), procurement status, and real-time field activities.There is an urgent need for a holistic, multi-stage system that can intelligently plan tasks, allocate resources, and adaptively respond to changes and constraints at all levels of a construction project.

[0003] The construction industry has long struggled with challenges in efficient scheduling and resource allocation, particularly for large and multi-stage projects such as high-rise buildings, infrastructure corridors, and industrial complexes. Traditional methods for managing construction processes typically rely on manual scheduling tools, spreadsheet-based planning systems, and general project management software like Microsoft Project or Primavera P6. These tools are inherently linear and static, offering minimal support for the dynamic realities of a construction site, where tasks are often interdependent, environmental conditions change rapidly, and resource availability fluctuates due to various factors such as work shifts, equipment breakdowns, and procurement delays.While these software solutions enable the formulation of a basic project plan and the tracking of progress using milestones, they are not sophisticated enough to automatically and adaptively reallocate resources or adjust the task sequence in real time in the event of deviations or constraints.

[0004] In modern construction environments, Building Information Modeling (BIM) systems have established themselves as a powerful digital platform for the three-dimensional management of architectural and engineering data. BIM enables stakeholders to visualize project components, analyze structural dependencies, and identify conflicts in the physical configuration of building elements. While BIM excels at design coordination and space planning, it is not inherently designed for dynamic scheduling or real-time resource management. Existing integrations between BIM and scheduling platforms are typically unidirectional: schedules are manually derived from BIM models but not dynamically adjusted to actual site conditions. This separation leads to delayed decisions, limited coordination between teams, and inefficiencies in the deployment of labor and equipment on site.

[0005] Several enterprise-level project management suites attempt to improve planning by integrating Earned Value Management (EVM), Key Performance Indicators (KPIs), and historical cost analysis modules. While these can provide valuable insights into project status and performance trends, they are reactive rather than proactive. These systems primarily function as monitoring tools; they do not automatically modify task sequences or reallocate resources to address bottlenecks. Furthermore, such systems often operate in isolation and cannot be integrated with on-site data collection mechanisms such as IoT sensors, machine telemetry, or GPS-based equipment tracking. This results in a critical information lag, where decisions are made based on outdated or incomplete data, leading to suboptimal allocation of personnel and material resources.

[0006] In recent years, mobile construction management platforms have also been introduced, enabling real-time communication between field teams and project managers. These solutions typically support checklist-based task tracking, image uploads, issue tagging, and basic time tracking. However, they rarely support intelligent workflow optimization or automated planning. Their usefulness is even more limited on large-scale projects with multiple vertical or horizontal construction zones, where coordinating teams working on different subsystems—electrical, plumbing, heating, ventilation, air conditioning, structural steelwork, etc.—requires a high degree of synchronization.Since these mobile platforms lack hierarchical task modeling and predictive sequencing, they are unsuitable for projects with multi-layered construction processes such as floor-by-floor concreting, vertical scaffolding systems, and interdependent inspection cycles.

[0007] Another class of solutions involves the use of event-driven simulation and project modeling software specifically tailored to the construction industry. These tools can model complex interactions between tasks, simulate resource constraints, and identify potential delays. However, their adoption is limited because precise simulation configuration requires a high level of technical expertise. They are often used in academic or consulting contexts rather than in real-time, on-site execution. Furthermore, these models are typically created during the pre-construction phase and not integrated into daily operations. Consequently, their adaptability to disruptions during operation or anomalies on the construction site is limited.

[0008] Automation and robotics are also used selectively in construction planning and execution. For example, autonomous robots are employed for tying reinforcing bars or masonry machines to address labor shortages and improve quality. While these technologies represent a step towards the mechanization of certain tasks, they operate in isolation and are not integrated into broader resource management. Their use remains task-specific and inadequately considers the surrounding construction context or the overall project schedule. Furthermore, their effectiveness is compromised without a coordination system that dynamically adapts robotic or human resources to changing project conditions.

[0009] There have also been attempts to integrate GPS and RFID technologies for equipment and personnel tracking. While such systems can provide real-time geolocation data, they are rarely used for active planning or resource optimization. For example, knowing that a crane is idle or a concrete mixer is unused does not automatically lead to a reassignment of these assets unless a human operator intervenes. Existing systems lack intelligent decision-making modules that correlate spatial data with task priority and reallocate resources in a cost-effective manner.

[0010] Cloud-based construction management platforms have recently gained popularity due to their ability to centralize documentation, communication, and workflow updates. However, these platforms often fall short when it comes to multi-level task decomposition and dynamic planning. They fail to reflect the hierarchical nature of construction projects, where macro-level tasks, such as completing the structural work, consist of several mid-level operations (e.g., beam placement, column pouring) and further micro-level tasks (e.g., formwork placement, reinforcement connection, concrete vibration). Without this decomposition, planning decisions are often too broad, leading to inefficient overlaps or gaps in task execution.

[0011] Perhaps the biggest shortcoming of existing solutions is the lack of integrated conflict resolution. Conflicts frequently arise on real-world construction sites—two teams need access to the same work platform, or one piece of equipment is scheduled to serve two zones simultaneously. Most current systems rely on manual monitoring to detect and resolve such conflicts, which is inefficient and prone to errors. Furthermore, they lack simulation-based predictive capabilities to anticipate future conflicts based on current decisions. The absence of automated impact analysis and redistribution logic means that even minor schedule deviations can lead to critical delays and cost overruns.

[0012] The construction technology landscape has evolved significantly with the introduction of BIM, cloud collaboration tools, mobile apps, and data tracking technologies. However, none of these systems fully addresses the need for intelligent, multi-stage workflow planning and adaptive resource allocation. Existing tools are either too general, too static, or too specialized to effectively coordinate complex, interdependent, and dynamically changing construction projects. The lack of integration between design data, real-time field conditions, and predictive design models results in a fragmented execution environment where inefficiencies increase and project performance suffers.There is a strong and unmet need for a unified, intelligent system that can model hierarchical construction processes, dynamically allocate resources, and adaptively resolve task conflicts in real time, thereby transforming construction project management from a reactive to a proactive and optimized discipline. Summary of the invention

[0013] The invention describes a system and an associated device for planning multi-stage construction processes and allocating construction resources. The system comprises a multi-stage architecture with a central planning engine (CSE), a distributed execution unit (DEU) at the respective locations, a resource coordination controller (RCL), and an adaptive conflict resolution unit (ACRM). The central planning engine receives input from architectural models (e.g., BIM), structural planning logic, procurement schedules, and real-time feedback from the DEUs. It generates a hierarchical schedule that includes project milestones at the macro level, structural dependencies at the mid-level (e.g., floor-by-floor sequencing), and task assignments at the micro level (e.g., welding, wiring, concreting).The resource coordination controller manages the availability and allocation of machinery, labor, security personnel, and materials, taking into account real-time constraints and predicted demand. The DEU units are machine-integrated hardware modules deployed across various site zones to execute schedule instructions, monitor progress via integrated sensors, and provide feedback to the CSE for iterative re-optimization. The ACRM module autonomously detects task conflicts, resource idle states, or bottlenecks and, based on an impact analysis, proposes rescheduling or resource reallocation.

[0014] The main objective of the present invention is to provide a comprehensive and intelligent system for managing multi-stage construction scheduling and resource allocation that overcomes the limitations of existing project management tools. The invention is designed to orchestrate construction activities across hierarchical task structures—from macro-level project milestones to micro-level field operations—by integrating real-time field data, architectural dependencies, and predictive design models into a unified decision-making framework. A further objective of the invention is to enable dynamic and adaptive resource allocation for construction workers, equipment, and materials based on constantly evolving site conditions, logistical constraints, and task dependencies. This minimizes idle time, resource conflicts, and execution delays.The invention also aims to provide an embedded, machine-integrated device for use on construction sites, functioning as a distributed execution node. It communicates directly with personnel and machinery on site, monitors task execution via sensors and telemetry, and makes planning decisions locally and synchronously with a central planner. A further objective of the invention is to introduce an autonomous conflict resolution mechanism that detects scheduling conflicts, evaluates mitigation strategies through real-time impact simulation, and automatically reorders tasks or reallocates resources to optimize construction continuity. The invention is also intended to improve communication, accountability, and project transparency by providing an integrated project dashboard that visualizes progress, detects deviations from the schedule, and delivers actionable insights for proactive decision-making.Ultimately, the invention aims to transform construction project management into an intelligent, data-driven process that dynamically adapts to the complexity and uncertainties of multi-stage construction environments. BRIEF DESCRIPTION OF THE FIGURE

[0015] These and other features, aspects, and advantages of the present invention will be better understood if the following detailed description is read with reference to the accompanying drawing, in which the same symbols consistently represent the same parts. The following applies: Fig. Figure 1 shows a block diagram of a system for multi-stage planning of construction processes and resource allocation.

[0016] Experts will also recognize that the elements in the drawing are shown for the sake of simplicity and are not necessarily to scale. For example, the flowcharts illustrate the process by highlighting the main steps to enhance understanding of the aspects of this disclosure. Furthermore, with regard to the design of the device, one or more components of the device may be represented in the drawing by conventional symbols, and the drawing may show only the specific details relevant to understanding the embodiments of this disclosure, so as not to clutter the drawing with details that are readily apparent to those skilled in the art after reading this description. Detailed description of the invention

[0017] For a better understanding of the inventive principles, reference is made below to the embodiment shown in the drawing, which is described in specific terminology. However, this does not limit the scope of the invention. Changes and further modifications of the illustrated system, as well as further applications of the inventive principles, are possible, as would normally occur to a person skilled in the art in the field of invention.

[0018] It is clear to the person skilled in the art that the preceding general description and the following detailed description are exemplary and explanatory of the invention and are not intended as a limitation of it.

[0019] References in this specification to “an aspect”, “another aspect”, or similar expressions mean that a particular feature, structure, or property described in connection with the embodiment is included in at least one embodiment of the present disclosure. Therefore, occurrences of the expressions “in one embodiment”, “in another embodiment”, and similar expressions in this specification may all refer to the same embodiment, but need not.

[0020] The terms "includes," "include," or other variations thereof are intended to cover non-exclusive inclusion, such that a process or method that includes a list of steps may not only contain those steps but may also include other steps not expressly listed or inherent in such process or method. Likewise, the statement "includes..." in the case of one or more devices, subsystems, elements, structures, or components does not, without further limitations, preclude the existence of other devices, subsystems, elements, structures, components, or additional devices, subsystems, elements, structures, or components.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by a person skilled in the art in the field of the invention. The system, methods, and examples provided here serve only for illustration and are not to be construed as a limitation.

[0022] Embodiments of the present disclosure are described in detail below with reference to the attached drawing.

[0023] Figure 100 shows a block diagram of a system for multi-stage planning of construction processes and resource allocation. System 100 comprises: a central planning engine (102) that receives input data such as architectural design models, structural constraints, procurement schedules, and historical performance indicators; a task decomposition processor (104) that is operationally linked to the central planning engine and generates a hierarchical construction task graph by decomposing macro-level construction milestones into mid-level and micro-level subtasks, with each subtask being provided with time estimates, location identifiers, resource requirements, and interdependencies;a hybrid planning technique (106) executed by the central planning engine, wherein the hybrid planning processing unit comprises a reinforcement learning component trained on historical construction sequence data and a constraint-based optimizer that resolves time and resource constraints across the entire task graph; a resource coordination layer (108) operationally connected to the central planning engine. The resource coordination controller comprises a real-time database of work units, machines, and material inventories, each resource having attributes such as availability, usage history, operational status, and spatial location;a plurality of distributed execution units (110) deployed in the construction zones, each distributed execution unit comprising an embedded controller, sensor interfaces, task status processing logic and communication circuits (110a), each distributed execution unit being configured to receive planning instructions from the central planning machine, execute localized control logic for task confirmation and resource activation, and transmit task execution data back to the central planning machine;an adaptive conflict resolution processing unit (112) that is operationally connected to the central scheduling engine and configured to detect conflicts in task execution or resource conflicts, simulate alternative task-resource allocation scenarios using a real-time multi-agent model, and autonomously update the task graph with revised task sequences and resource allocations; and a construction workflow dashboard (114) configured to visualize task progress, deviations from the planned schedule, and resource efficiency metrics, the dashboard also being capable of receiving manual override inputs or approving automated conflict resolution proposals generated by the adaptive conflict resolution module.

[0024] In one embodiment, the task decomposition processor (104) also includes a semantic parser configured to extract task metadata from Building Information Modeling (BIM) files using natural language processing and object recognition techniques, thereby automatically assigning spatial components in the BIM model to the corresponding subtasks in the construction sequence.

[0025] In one embodiment, the hybrid scheduling processing unit (106) comprises a genetic engineering component configured to generate an initial feasible solution set for the task graph under multi-objective constraints, including minimum throughput time, critical path compression, and resource leveling, with subsequent training of reinforcement learning agents to refine these solutions based on simulation feedback from completed projects.

[0026] In one embodiment, each distributed execution unit (110) comprises: a machine vision module configured to detect physical task states using image recognition of tools, structures, and the worker's posture; a tool telemetry interface configured to receive vibration, load, or utilization signals from connected machines; and an edge inference engine that executes lightweight models for local task validation, thus enabling autonomous confirmation of task completion prior to updating the central scheduling engine.

[0027] In one embodiment, the adaptive conflict resolution processing unit (112) simulates alternative execution scenarios using a graph-based causal inference model, wherein each task node is provided with probabilistic execution windows and dynamic dependencies derived from temporal uncertainty models, thereby enabling proactive reordering to minimize the probability of downstream conflicts.

[0028] In one embodiment, the resource coordination controller (108) also includes an availability forecasting engine that estimates future resource availability using a recurrent neural network trained on historical attendance logs, maintenance cycles, and delivery records, thus enabling proactive resource reservation for critical upcoming tasks.

[0029] In one embodiment, the central scheduling engine (102) comprises a latency-aware scheduling module configured to take into account communication delays between zones and variations in execution latency in distributed execution units, adjusting task assignment decisions to account for computation delays at the edges and network jitter tolerances.

[0030] In one embodiment, the dashboard (114) for the construction process also includes a 4D visualization interface that overlays temporal progress metrics onto a 3D representation of the project structure, with each building element being color-coded based on the task completion status and the efficiency of resource utilization.

[0031] In one embodiment, the communication circuit (110a) of each distributed execution unit comprises a multiprotocol transceiver that can be operated via LoRaWAN, LTE and Wi-Fi interfaces, and in which a fallback synchronization mechanism is implemented to ensure the buffering and playback of task data under conditions of intermittent connectivity.

[0032] In one embodiment, the task graph generated by the task decomposition processor is encoded in a directed acyclic graph structure (DAG), with a task priority encoder calculating the task weights based on a composite function of structural criticality, schedule margin, and dependency centrality.

[0033] The system for multi-stage construction scheduling and resource allocation integrates a sophisticated computer architecture that orchestrates task planning, optimizes resource utilization, and dynamically adapts to site variability in real time. At the heart of the system is the central scheduling engine, which executes a hybrid scheduling processing unit combining elements of reinforcement learning, genetic optimization, and constraint-based task sequencing. This technology is initialized with project-specific inputs such as architectural models, engineering constraints, procurement plans, and historical project data. A task decomposition processor integrated into the scheduling engine performs a multi-stage decomposition of project milestones into granular subtasks by analyzing Building Information Modeling (BIM) files and associated metadata.Using natural language processing and object recognition techniques, this module identifies spatial and temporal elements such as floors, structural segments, or mechanical systems and assigns them to feasible construction activities. The resulting construction task graph is structured as a directed acyclic graph (DAG), where each node of a task and each edge represents a dependency or precedence condition.

[0034] The processing unit for hybrid planning begins by generating an initial feasible schedule using a multi-objective genetic technique. This technique searches the solution space using crossover, mutation, and fitness assessment steps, optimizing for goals such as minimal overall throughput time, even resource leveling, minimal idle time, and priority-based task sequencing. The fitness function also considers weighted penalty points for violations of interdependencies or resource constraints. The result is a near-optimal baseline plan, which is then further refined by a reinforcement learning agent.The agent is trained using simulation data from previous projects and is designed to adapt to dynamic field conditions by learning optimal action guidelines for reordering tasks, reallocating resources, and adjusting timing in response to disturbances such as resource unavailability, task delays, or environmental influences.

[0035] In real-time provisioning, the technology operates with a rolling horizon. The entire construction schedule is divided into time segments or "forecast windows" that are continuously updated. Within each window, the task graph is re-evaluated based on current resource availability and feedback on task execution. If the system receives a deviation signal—for example, a delay in formwork removal or the unavailability of a crane—the reinforcement leasing agent checks alternative execution paths and updates the schedule locally without requiring a complete recalculation. This modularity ensures that the planning system remains computationally efficient even in complex, large-scale construction environments.

[0036] The resource coordination controller acts as a live repository of all available workers, machines, and materials. Each resource entry is linked to metadata attributes such as geolocation, operational status, expected availability window, maintenance history, and compatibility with task requirements. These attributes are updated in real time via telemetry data from distributed execution units and enterprise resource planning (ERP) integrations. A predictive availability engine, based on a recurrent neural network, forecasts the expected resource availability in upcoming planning windows. This predictive capability allows the system to reserve critical resources in advance and proactively redirect tasks to alternative teams or equipment where possible.

[0037] The distributed execution units deployed on the construction site serve as edge nodes for task execution. Each unit is equipped with an embedded processor, an image processing system, LiDAR, RFID, and a range of environmental and operational sensors. These units receive planned task packages from the central planning engine and autonomously verify task readiness by validating the task prerequisites using sensor data. For example, a distributed execution unit monitoring a concreting operation can visually confirm the placement of the reinforcement, scan RFID tags to verify the formwork position, and analyze vibration data from mixers to ensure operational status. A local inference engine on the edge device evaluates this multimodal sensor data to autonomously determine whether the prerequisites for execution are met.After confirmation, the unit can activate connected machines or assign the task to a specific authenticated worker using biometric verification.

[0038] A particularly important component of the system is the adaptive conflict resolution module. This module continuously monitors the task graph for potential collisions, such as when two teams require access to the same physical location or a single resource is overbooked. It utilizes a simulation-based engine that models alternative task sequences using multi-agent graph simulation. Each task and resource is modeled as an intelligent agent with defined states and permissible transitions. The simulation iteratively examines possible alternative plans, quantifies trade-offs using cost-impact models, and outputs a range of potential remedial measures. These can include temporary task reassignments, substituting equivalent resources, or spatially reassigning activities.The module then either autonomously implements the most efficient scenario or presents options via the construction process dashboard for management approval.

[0039] The system also ensures compliance with safety protocols and sequencing logic through safety-aware scheduling restrictions integrated into the scheduling engine. For example, if a welding operation is performed in an area where combustible insulation has not yet been removed, the distributed execution unit detects the hazard using thermal or optical sensors and blocks the activation of the welding system. These restrictions are enforced by a real-time safety rule engine integrated into the scheduling logic to prevent unsafe task starts.

[0040] The construction process dashboard provides human stakeholders with a central interface for visualizing the execution status. The dashboard can display a 4D model (3D + time) of the project overlaid with real-time task completion data and resource efficiency metrics. It allows users to analyze individual task nodes, view historical execution logs, and override automated decisions as needed. The dashboard also includes an anomaly detection module that uses unsupervised clustering to identify patterns indicating systemic inefficiencies, such as recurring task delays or underutilized machinery. These anomalies are flagged and highlighted in the interface to support corrective action.

[0041] By integrating the aforementioned technical and architectural elements, the system enables intelligent, responsive, and scalable coordination of construction processes. It dynamically adapts to the execution conditions, optimizes resource allocation in detail, and minimizes project delays through predictive and autonomous planning. This eliminates long-standing inefficiencies in the execution of complex construction projects.

[0042] The multi-stage system for scheduling construction processes and allocating resources described here comprises several interacting modules organized within a unified orchestration pipeline. At its core is the Central Scheduling Engine (CSE), a software module that runs on a cloud platform or a dedicated on-premises server. The CSE processes structural and architectural data files, including 3D BIM models, 2D blueprint PDFs, and dependency diagrams from structural engineering analysis. From this input, the CSE generates a task decomposition matrix that categorizes each construction deliverable into a hierarchy of subtasks characterized by spatial location, resource requirements, interdependencies, and time estimates. This decomposition is then subjected to constraint-based optimization.This involves using a hybrid model that combines genetic techniques and Critical Path Method (CPM) planning and is extended by reinforcement learning agents that learn optimal sequencing patterns from historical project data.

[0043] The system also includes a Resource Coordination Controller (RCL) that manages a dynamic database of available construction resources, including skilled and unskilled workers, scaffolding and cranes, concrete mixers, power tools, and safety equipment. Each resource is tagged with metadata, including availability window, compatibility with task type, geolocation, and usage history. The CSE communicates with the RCL to dynamically assign resources to tasks, ensuring that the optimal resources are allocated to the appropriate tasks at any given time.

[0044] Several Distributed Execution Units (DEUs) are physically distributed across the construction site areas. Each DEU is a ruggedized embedded device with a multi-core processor, field I / O interfaces, RFID scanners, an edge AI chip, a camera module, LiDAR sensors, and a wireless communication unit. These DEUs receive task packets from the CSE, execute local scheduling logic to assign jobs to workers or activate machinery, and capture task status updates using image recognition and tool telemetry. For example, a DEU on the third-floor concrete pouring area can detect that a mixer has completed its batch cycle, verify reinforcement placement via image recognition, and authorize the activation of the concrete pump.

[0045] To manage resource conflicts and dynamic disruptions such as weather, labor shortages, or machine breakdowns, the system features an Adaptive Conflict Resolution (ACRM) unit. The ACRM performs a real-time simulation of task execution using a multi-agent model, where each task and resource is represented as a decision node. If a conflict arises—for example, when two teams require access to the same scaffolding—the ACRM conducts an impact analysis, examines alternative task reassignment scenarios, and proposes remediation strategies, such as deploying a different team, moving the task to an adjacent zone, or delaying the task to a minimal cost window.

[0046] Furthermore, the system integrates a Project Performance Dashboard (PPD) that visualizes progress indicators in real time, including performance metrics, deviations from planned schedules, machine idle rates, and task completion indices. The PPD allows project managers to intervene manually or approve reconfigurations generated by ACRM.

[0047] The invention relates to the field of construction project management systems, and in particular to intelligent planning and resource allocation systems for construction environments. Specifically, it is a computer-implemented and machine-integrated system for hierarchical workflow orchestration in multi-stage construction projects, encompassing real-time task monitoring, predictive resource planning, and adaptive conflict resolution using advanced artificial intelligence, embedded hardware controllers, and edge computing. The invention overcomes the limitations of existing static planning tools by enabling dynamic, context-sensitive task sequencing and autonomous decision-making across distributed construction zones.

[0048] The drawing and the preceding description show examples of embodiments. Those skilled in the art will recognize that one or more of the described elements can be combined to form a single functional element. Alternatively, certain elements can be divided into several functional elements. Elements of one embodiment can be added to another embodiment. For example, the sequence of the processes described here can be changed and is not limited to the manner described here. Furthermore, the actions of a flowchart need not be implemented in the sequence shown; nor does it necessarily have to be performed by all actions. Actions that are not dependent on other actions can also be performed in parallel with the other actions. The scope of the embodiments is in no way limited by these specific examples.Numerous variations are possible, whether explicitly stated in the specification or not, such as differences in structure, dimensions, and material use. The range of embodiments is at least as broad as specified in the following claims.

[0049] Advantages, further benefits, and problem solutions have been described above with reference to specific embodiments. However, the advantages, benefits, problem solutions, and all components that can lead to an advantage, benefit, or solution occurring or becoming more apparent are not to be construed as critical, necessary, or essential features or components of individual or all claims. REFERENCES 100 A system for the multi-stage planning of construction processes and the allocation of resources. 102 Central Planning Machine 104 Task Decomposition Module 106 Hybrid Scheduling Processing Unit 108 Resource Coordination Controllers 110 Distributed Execution Units 110a Communication circuit 112 Adaptive Conflict Resolution Processing Unit 114 Dashboard for the Construction Process

Claims

[1] A system for multi-stage planning of construction processes and resource allocation, consisting of: a central planning engine configured to receive input data, including architectural design models, structural constraints, procurement schedules, and historical performance indicators; a task decomposition processor that is operationally connected to the central planning engine and configured to generate a hierarchical construction task graph by decomposing macro-level construction milestones into mid-level and micro-level subtasks, with each subtask having time estimates, location identifiers, resource requirements, and mutual dependencies; a hybrid planning processing unit configured to resolve time and resource constraints across the entire task diagram; a resource coordination controller that is operationally connected to the central planning engine, wherein the resource coordination controller includes a real-time database of work units, machines and material stocks, each resource being tagged with attributes such as availability, usage history, operating status and spatial location; a multitude of distributed execution units distributed across the construction zones, each distributed execution unit comprising an embedded controller, sensor interfaces, task status processing logic, and communication circuitry, each distributed execution unit being configured to receive planning instructions from the central planning machine, execute localized control logic for task confirmation and resource activation, and transmit task execution data back to the central planning machine; an adaptive conflict resolution processing unit that is operationally connected to the central planning engine and configured to detect conflicts in task execution or resource conflicts, simulate alternative task-resource allocation scenarios using a real-time multi-agent model, and autonomously update the task graph with revised task sequences and resource allocations; and A dashboard for the construction process, configured to visualize task progress, deviations from the planned schedule, and resource efficiency metrics, with the dashboard also being able to receive manual override inputs or approve automated conflict resolution proposals generated by the adaptive conflict resolution module. [2] System according to claim 1, wherein the task decomposition processor further comprises a semantic parser configured to extract task metadata from Building Information Modeling (BIM) files using natural language processing and object recognition techniques, thereby automatically assigning spatial components in the BIM model to the corresponding subtasks in the construction process. [3] System according to claim 1, wherein the hybrid planning processing unit includes a genetic engineering component configured to generate an initial feasible solution set for the task graph under multi-objective constraints, including minimum throughput time, critical path compression and resource leveling, with subsequent reinforcement learning agents trained to refine these solutions based on simulation feedback from completed projects. [4] System according to claim 1, wherein each distributed execution unit comprises: a machine vision module configured to detect physical task states using image recognition of tools, structures, and the worker's posture; a tool telemetry interface configured to receive vibration, load, or usage signals from connected machines; and an edge inference engine that runs lightweight models for local task validation, thus enabling autonomous confirmation of task completion before the central planning engine is updated. [5] System according to claim 1, wherein the adaptive conflict resolution processing unit simulates alternative execution scenarios using a graph-based causal inference model, wherein each task node is provided with probabilistic execution windows and dynamic dependencies derived from temporal uncertainty models, thereby enabling proactive reordering to minimize the probability of downstream conflicts. [6] System according to claim 1, wherein the resource coordination controller further comprises an availability forecasting engine which estimates future resource availability using a recurrent neural network trained on historical attendance logs, maintenance cycles and delivery records, thus enabling predictive resource reservation for critical upcoming tasks. [7] System according to claim 1, wherein the central planning machine comprises a latency-aware planning module configured to take into account delays in communication between zones and variations in execution latency in distributed execution units, adapting task assignment decisions to account for delays in boundary computation and network jitter tolerances. [8] System according to claim 1, wherein the dashboard for the design process further comprises a 4D visualization interface that overlays temporal progress metrics onto a 3D representation of the project structure, wherein each design element is color-coded based on the task completion status and the efficiency of resource use. [9] System according to claim 1, wherein the communication circuit of each distributed execution unit comprises a multiprotocol transceiver that can be operated via LoRaWAN, LTE and Wi-Fi interfaces, and wherein a fallback synchronization mechanism is implemented to ensure the buffering and playback of task data under conditions of intermittent connectivity. [10] System according to claim 1, wherein the task graph generated by the task decomposition processor is encoded in a directed acyclic graph structure (DAG) and wherein a task priority encoder computes task weights based on a composite function of structural criticality, schedule margin and dependency centrality.

Images