Alerta

一种智慧物流园区与仓储决策方法及系统

Absstract of: CN121745810A

本发明提供一种智慧物流园区与仓储决策方法及系统，涉及物流仓储技术领域，本发明通过实时监测运输机构位置，在检测到多个运输机构共用同一货架节点时，将其作为冲突节点，基于时间裕度确定执行优先级，并分别计算额外等待时间成本和改道决策时间成本，最终根据时间裕度与时间成本的比较结果动态决策执行顺序与策略，实现冲突节点处的协同调度与全局优化，提升物流效率与订单交付及时性。

一种加氢站设备完整性知识图谱应用系统

Absstract of: CN121745242A

本发明公开了一种加氢站设备完整性知识图谱应用系统及方法，属于氢能基础设施全生命周期管理技术领域，其工艺数据采集与定义模块用于数据采集和定义，工艺知识抽取模块用于从工艺文档中抽取包含工艺环节、工艺参数、标准值的三元组及从操作记录中抽取关于工艺环节、执行人员、时间的关联关系，并建立工艺环节与设备、工艺参数异常与设备故障的关联关系；知识图谱架构储存模块用于构建并储存多层级知识图谱架构，多层级知识图谱架构包括基础层、工艺层、实体层、关系规则层、设备三维图层、风险管理决策层。本发明解决了现有加氢站设备完整性知识图谱缺乏工艺维度整合、设备与工艺协同管理能力弱的问题。

基于双层车辆路径优化的跨接卸货与同时取送货物流路径优化方法

Absstract of: CN121745428A

本发明属于物流路径优化技术领域，公开了一种基于双层车辆路径优化的跨接卸货与同时取送货物流路径优化方法，包括：将以旧换新物流中的双层车辆路径规划构建为三级物流网络的车辆路径规划；将三级物流网络的车辆路径规划中所涉及的节点和节点间的路径构建为完全有向图，进而构建双层车辆路径规划对应的混合整数线性规划模型MILP，以最小化总运输成本为目标函数，以包括客户货物在指定时间窗内完成的时间窗约束，车辆容量约束以及双层车辆路径的同步约束在内的约束下，求解MILP，得到双层车辆路径规划方案。还提供了一种自适应禁忌搜索算法，结合导向型扰动机制和阈值加速策略，解决了多层级网络协调、动态约束适应和大规模实例计算效率的挑战。

一种废钢企业集中管控大屏系统

Absstract of: CN121745840A

本发明公开了一种废钢企业集中管控大屏系统，属于废钢信息化技术领域。本发明包括生产经营监控分析大屏模块、业务运营管控追踪大屏模块、反向开票监控单元、动态地图单元、场内运行监控单元及数据加密子模块。本发明通过双大屏协同架构实现核心经营指标动态计算、吨钢节约能源自动分析、采购销售多维度可视化；创新构建经营规模风险修正模型，动态计算销售额/纳税额/能源节约量；建立反向开票高亮预警机制与物流‑台账联动模型，支持15分钟级业务刷新；通过TLS加密传输与权限分级控制，确保经营数据安全性与决策实时性，显著提升废钢企业资源管控精度与供应链协同效率。

基于区块链的道路施工设备调度方法及系统

Absstract of: CN121745540A

本申请涉及道路施工的技术领域，公开基于区块链的道路施工设备调度方法及系统，该道路施工设备调度方法包括以下步骤：采集每台施工设备的运行状态、地理位置及部件寿命数据，并加密上链，得到区块链数字镜像；采用自适应多目标分析算法分析设备利用率与施工进度的关联，通过模拟不同调度方案的效果以生成优化调度方案；将优化调度方案转化为智能合约，在满足施工预设条件时，自动触发并执行设备调度与运输任务派发；识别进度偏差或位置偏差；根据识别的进度偏差或位置偏差，生成新的优化调度方案。本申请在每次迭代中动态调整调度方案，确保设备的利用率和施工进度最大化，有效应对复杂的施工现场环境。

一种海铁联运港口作业区布局优化方法

Absstract of: CN121745355A

本公开实施例中提供了一种海铁联运港口作业区布局优化方法，属于计算技术领域，具体包括：确定待规划区域和待设置的作业区；将待规划区域和该区域内待设置作业区简化为矩形，以待规划区域左下角为原点建立平面直角坐标系，将各作业区的中心点坐标作为布局控制点；根据货物流通数据分析待规划区域内各作业区之间的物流关系、非物流关系和综合相关程度，确定货物搬运距离；建立目标函数，引入归一化因子及权重系数，将多目标函数转化成单目标函数，建立作业区布局约束条件，据此构建海铁联运港口作业区布局规划模型；采用改进鲸鱼优化算法迭代求解海铁联运港口作业区布局规划模型。通过本公开的方案，提高了优化效率、精准度和适应性。

一种在智慧物流中基于MARL的实时任务调度优化方法

Absstract of: CN121745785A

本发明公开一种在智慧物流中基于MARL的实时任务调度优化方法，包括以下步骤：1）确定智慧物流场景下的系统结构，并构建蜂窝用户设备CUE、车载用户设备VUE的任务时延表达式；2）基于任务时延表达式，构建实时时延保障优化模型；3）将VUE建模为分布式智能体，并通过多智能体强化学习方法求解实时时延保障优化模型，得到实时物流任务调度方案。本发明通过近端策略优化（Proximal Policy Optimization，PPO）算法实现本地策略迭代，结合联邦学习框架进行全局参数聚合，有效突破传统集中式优化的信息孤岛瓶颈。

一种基于深度强化学习的需求驱动物料需求计划系统及优化方法

Absstract of: CN121745557A

本发明属于智能制造技术领域，其公开了一种基于深度强化学习的需求驱动物料需求计划系统及优化方法；该系统包括需求模块，用于采集处理并输出相关需求与供应商环境数据；需求驱动物料需求计划模块，用于生成库存补货计划、生产计划与物料采购计划；仿真模块，用于构建并运行仿真模型，以输出仿真实验统计结果；深度强化学习模块，与需求驱动物料需求计划模块和仿真模块进行交互，利用内置算法库进行训练，输出需求驱动物料需求计划的优化参数集；执行模块，用于将计划信息转化为作业指令并反馈实际运作数据与执行结果，以弥补传统需求驱动物料需求计划静态参数配置的不足，实现库存水平与服务率的协同优化，增强企业对市场需求变化的响应能力。

一种装配式金属结构智能验收与溯源系统

Absstract of: CN121745732A

本发明公开了一种装配式金属结构智能验收与溯源系统，包括应变监测模块、缺陷检测模块、数据采集模块、溯源管理模块、智能分析模块和存证模块。通过应变敏感二维码实现低成本应力监测，多光谱融合技术进行缺陷检测，数字指纹技术建立质量追溯，区块链技术保证数据不可篡改。系统采用事件驱动机制协同工作，从结构安全性、材料完整性、工艺规范性三个维度评估质量，通过集成学习生成验收决策。本发明解决了现有装配式金属结构验收中应力监测成本高、缺陷检测不全面、质量追溯困难的问题，为装配式建筑质量管理提供了系统化解决方案。

一种基于人工智能(AI)的库存补货方法

Absstract of: CN121745819A

本系统是一种基于人工智能(AI)的库存补货解决方案，旨在通过自动化和智能化的技术手段，优化企业的库存管理流程。系统通过集成多种数据源、利用先进的AI模型进行预测与决策，为企业提供精准的库存补货策略，确保库存的高效管理，减少缺货和过度库存的风险。是现代供应链管理中的关键组成部分，旨在通过智能分析历史销售数据、市场需求预测、供应链效率优化等手段，自动化地调整库存水平，以减少库存积压、避免缺货情况，并提高整体运营效率。该系统通过智能化的库存管理和补货决策，大幅提高库存管理效率，降低库存成本，并提升客户满意度。它适用于各种规模的企业，尤其适合供应链复杂、需求波动大的行业，如零售、制造和电子商务。

一种面向无人车无人机联动的物流配送调度方法及系统

Absstract of: CN121745794A

本申请涉及一种面向无人车无人机联动的物流配送调度方法及系统。所述方法包括：采集风速时序数据、道路拥堵影像、建筑三维点云等多模态实时数据，经跨模态特征融合技术生成动态环境条件与静态特征描述；基于核心特征通过加权决策树算法确定初步配送模式，无人机倾向时提取路径关联特征并结合实时风速计算风险系数，生成路径可行性评估结果；依据风险系数确定最终配送模式，通过A星算法生成优化路径，结合拥堵预测与实时天气核算耗时，输出符合时效的配送时间安排；基于传感器反馈数据计算位置与时间偏差，超阈值时生成更新路径并下发调整指令，监控执行至目的地后生成任务完成确认信息。本方法提升了联动物流配送调度的精准性、安全性与高效性。

一种升降横移立库入库区域的选择系统

Absstract of: CN121738407A

本申请提供一种升降横移立库入库区域的选择系统，该选择系统可以根据用户的取车时间预约，主动将计划在某一集中时间段取车的用户引导至不同的独立运营区域。这使得取车需求在时间上被分散，避免了所有用户在相近时间点集中于同一区域取车，极大地缓解了高峰期拥堵，提升了整个立库的吞吐效率和设备运行效率，无需更多数量的独立运营出入库的停车区域来实现高峰避让，实现错峰调度，显著提升效率。在新建车库时，可以提前根据该系统的数据统计有效的减少独立运营出入库的停车区域数量，从而可以减少避让空间以增加停车车位。用户通过预约取车时间，可以获得系统推荐的最佳停车区域。

确定物流的中转用户的方法、装置、电子设备和存储介质

Absstract of: CN121745783A

本公开提供一种确定物流的中转用户的方法、装置、电子设备和存储介质，涉及物流技术领域。上述方法包括：获取目标用户在物流过程中的中转相关特征，所述中转相关特征包括收件特征和寄件特征；调用中转用户判断模型处理所述中转相关信息，得到所述目标用户的中转判断信息，所述中转判断信息用于指示所述目标用户是否为中转用户。本公开通过中转用户判断模型，判断出物流的中转用户，提高了判断的准确度和稳定性。并且，参考用户在物流过程中的寄件行为的特征和收件行为的特征，可以全面、综合地判断用户是否为中转用户。

自动负载装卸装置及系统

Absstract of: WO2025046544A1

A material handling system (10, 110, 210) includes a storage apparatus (12, 112, 212) having a frame (20, 120, 220) and a plurality of storage shelves, with the frame (20, 120, 220) configured to support the plurality of storage shelves on a floor, and an automated load handling device (14, 114, 214) having a base (28) configured to be a supported on and for movement across the floor, and the automated load handling device (14, 114, 214) including a stabilizing member (18, 118, 218) configured to engage the storage apparatus (12, 112, 212) to stabilize the automated load handling device (14, 114, 214).

基于增强现实的船舶危险货物异常识别方法及系统

Absstract of: CN121744159A

本申请提供了基于增强现实的船舶危险货物异常识别方法及系统，涉及增强现实技术领域，方法包括：构建多模态伴随异常空间；实时获取危险货物监测数据集，并进行显性异常识别；基于同一监测时区进行隐性异常派生优化；根据实时航行状态数据对显隐异常耦合图网进行航行扰动感知强化处理；对货物异常强化图网进行自适应动态风险评估，建立危险货物异常热力图，结合增强现实技术执行增强警示与处置导引。通过本申请解决了现有技术中由于危险货物异常识别方法主要依赖单证比对和显性规则，缺乏细粒度的风险评估，导致船舶危险货物异常识别不全面，进而影响船舶运输安全的技术问题，全面识别船舶危险货物异常，提高了船舶运输的安全性。

一种葡萄保鲜用气调库智能监控系统

Absstract of: CN121747031A

本发明涉及生鲜保鲜技术领域，且公开了一种葡萄保鲜用气调库智能监控系统，包括数据单元、图像单元、分离单元、处理单元、灰度单元、计算单元、中心单元以及管理单元，所述数据单元用于采集气调库内的环境数据，所述图像单元用于采集气调库内葡萄的图像数据，所述计算单元用于计算出体积变化值TB与灰度变化值HB，所述中心单元接收计算单元以及数据单元并进行预测值YC的计算与识别判断，本发明将环境数据与图像数据的多源数据融合，为智能监测提供基础，环境数据可以对葡萄储存的状态进行了解，图像数据对葡萄自身状态进行了解，根据以上两种类型的数据，所计算出的预测值YC可以对葡萄进行精确预测，进而实现葡萄保鲜的智能监控。

一种动态配额纠偏的物料需求计算方法、系统、设备及存储介质

Absstract of: CN121745627A

本发明公开了资源计划与供应链管理领域的一种动态配额纠偏的物料需求计算方法、系统、设备及存储介质，旨在解决传统MRP中配额固定、难以适应供应链动态变化及复杂业务场景的问题。所述方法包括：基于物料独立需求、BOM及成套组信息逐层分解物料需求；使用时序化BOM模型结合提前期展开需求并确定各周期毛需求；计算净需求并进行批量调整，形成建议采购数量；依据供应商历史绩效动态调整配额比例，并结合客户指定等约束分配至各供应商。本发明应用于工程机械制造业，能够实现配额动态适应、计划闭环优化与复杂场景自动处理，提升供应链韧性、库存周转与计划准确性。

ARTIFICIAL INTELLIGENCE AGENT USING LANGUAGE MODEL AND REINFORCEMENT LEARNING MODEL TO GUIDE PICKING PROCESS

Absstract of: WO2026064129A1

An artificial intelligence (Al) agent is disclosed that assists an entity to complete a task. The entity is assigned to complete a task. The Al agent monitors events to detect an occurrence of an event associated with the task. A machine learning model of the Al agent is prompted to generate a set of candidate actions based in part on the detected event and data about the entity. A reinforcement learning model of the Al agent scores each candidate action from the set to tailor the candidate actions to the entity. A scored action is selected as a recommended response to the event and is communicated to a client device of the entity which causes the entity to perform the selected action.

METHOD FOR DETERMINING A CONFIGURATION FOR PLACING AT LEAST ONE ITEM WITHIN A CONTAINER

Absstract of: WO2026061795A1

The invention relates to a method for determining a configuration for placing at least one item (ART) within a container (CNT), the method being implemented by an electronic device during order preparation, at an order-preparation station, for items to be placed within the container (CNT), the method comprising at least one iteration of the following steps: - capturing (S01) a datum representative of an available volume (VD) within the container (CNT); - obtaining (S04) a list (L1) of items to be placed within the container (CNT); - determining (S05) a volume (V1) of the items to be placed according to the list (L1) of the items to be placed; - depending on the volume (V1) of the items and on the datum representative of an available volume (VD) within the container (CNT), calculating (S06) the configuration (confP) for placing the items of the list (L1) of the items.

AUTOMATED DEVICE FOR CROSS-BORDER E-COMMERCE AND METHOD FOR CONTROLLING SAME

Absstract of: WO2026063564A1

An automated device for cross-border e-commerce and a method for controlling same are disclosed. The automated device according to one embodiment of the present invention comprises: a database that stores data; and a processor that controls the automated device, wherein the processor may, when a preset word representing a popular search term is identified while scraping a web page, extract keywords searched at a preset frequency or higher and store the keywords in the database, search a shopping platform using the extracted and stored keywords, store a shopping mall URL address searched on the shopping platform in the database, and store product data of the shopping mall in the database using the stored shopping mall URL address.

SYSTEM AND METHOD FOR SECURE PRODUCT TRACKING AND AUTHENTICATING USING GEO-TAGGING AND MACHINE-READABLE CODE INTEGRATION

Absstract of: WO2026062473A1

A system for secure product tracking and authentication using geo-tagging and machine-readable code integration is disclosed. The system includes a first computing device comprises a geo-tag and code generation module that creates a unique geo-tag for each product, embedding authorized data such as licenses, permissions, and trademark registrations. This geo-tag is paired with a machine-readable code containing product-specific details and a time-stamp, which is then embedded onto the product's packaging by a connected packing machine. A second computing device allows users to track the product across the supply chain, displaying real-time information on the product's location, status, and authenticity. A server manages and analyzes this data in real-time, ensuring regulatory compliance and detecting potential tampering. A scanning device reads the machine-readable code at various checkpoints, updating the product's status and transmitting data to the server, where alerts and notifications are generated based on analyzed data.

CONTROL DEVICE FOR CONTROLLING DISPLAY DEVICE PROVIDED AT SEAT

Absstract of: US20260087569A1

A control device controls a display device provided at a seat, including a processor and memory storing instructions that cause the control device to display an order screen on the display device, which is used by a customer seated in the seat to order items. The order screen includes a moving display area that sequentially displays each of multiple images, including images of items available for ordering, moving in a predetermined direction along a lane. The instructions further cause the control device to accept settings of items subject to discount and the remaining quantity of such discounted items, and display images announcing the discounted items and images indicating the remaining quantity of discounted items as part of the multiple images in the moving display area of the order screen, when settings for the item subject to discount and the remaining quantity of the discounted item are received.

Deep Causal Learning for Continuous Testing, Diagnosis, and Optimization

Absstract of: US20260086514A1

A system and methods for multivariant learning and optimization repeatedly generate self-organized experimental units (SOEUs) based on the one or more assumptions for a randomized multivariate comparison of process decisions to be provided to users of a system. The SOEUs are injected into the system to generate quantified inferences about the process decisions. Responsive to injecting the SOEUs, at least one confidence interval is identified within the quantified inferences, and the SOEUs are iteratively modified based on the at least one confidence interval to identify at least one causal interaction of the process decisions within the system. The causal interaction can be used for testing, diagnosis, and optimization of the system performance.

SELF-POSITIONING OF ELECTRONIC SHELF LABEL (ESL) DEVICES THROUGH INTER-ESL LINKS

Absstract of: US20260086186A1

This disclosure provides systems, methods, and devices for Electronic Shelf Label (ESL) systems that support positioning of ESL devices. In a first aspect, a method includes: determining known positions of a first subset of ESL devices; receiving positioning measurements for each ESL device in a second subset of ESL devices; estimating a first position of each ESL device in the second subset based on the positioning measurements received for that ESL device and the known positions of the first subset; determining a positioning error associated with the first position of each ESL device in the second subset; and estimating a second position of each ESL device in the second subset based on the positioning error and additional positioning measurements received for that ESL device. Other aspects and features are also claimed and described.

SYSTEM AND METHODS FOR ASSET TRACKING, ASSET GROUPING, AND ERROR RECOVERY

Nº publicación: US20260089469A1 26/03/2026

Applicant:

ZAINAR INC [US]
ZaiNar, Inc

Absstract of: US20260089469A1

A method includes, generating a schedule for an asset tag group, the schedule defining: a multicast trigger time for transmission of a multicast trigger to the asset tag group; and for each asset tag in the asset tag group, a transmit time succeeding receipt of the multicast trigger and unique to the asset tag, and a wake window intersecting the transmit time and the receipt of the multiact trigger. The method also includes, at the node network: broadcasting the multicast trigger; and, at an asset tag in the asset tag group, entering a wake mode; receiving the multicast trigger; transmitting a ranging signal; and entering the sleep mode. The method further includes, at the node network: receiving ranging signals from the asset tag group; and deriving locations of the asset tags in the asset tag group based on the ranging signals received by the node network.