Systems and methods generate and use an inventory map from a photograph of at least one physical object within a physical space. A server receives a photograph captured by a client device, parses the photograph to decode identification markers associated with an object, a container that holds the object, or a location indicator, and creates or updates an inventory record with a hierarchical location subrecord (e.g., address, building, room or space, shelf or collection space, container). When accompanying sensor data (orientation, positioning, inertial, depth) is available, the system determines and stores three-dimensional position attributes for the object within the container or space. A space component produces a two-dimensional plan and/or a three-dimensional view that places graphical representations according to the stored associations and positions. In some implementations, the system requests assignment of physical storage space from a storage provider and updates the inventory record with allocation results.

Disclosed are various embodiments for product blending optimization. In one embodiment, inventory data is received that represents a plurality of product layers stored in a plurality of storage units. The inventory data indicates a product quantity for each of the product layers and a set of characteristics for each of the product layers. Order data is received that represents a plurality of orders for a blended product. Each of the orders specifies a respective set of target characteristics for the blended product and a respective quantity. Based on the inventory data and the order data, an optimal sequence of the plurality of orders for fulfillment is automatically generated, along with an optimal sequence for dispensing product from the storage units for the blended product of each respective order.

Approaches for automated batch recall assessment are described. The approach includes identifying product batches having a plurality of faulty products manufactured by the organization. For each of the identified product batches, a plurality of quality concerns raised for faulty products manufactured as part of the product batch are quantified. Accordingly, for each product batch, the quantified values of each of the plurality of quality concerns is compared with a corresponding pre-determined threshold count value to enable determination of a quality risk level associated with the product batch. Based on the quality risk level, a batch recall assessment is performed to determine whether to recall product batches having the plurality of faulty products. Thus, the described approaches provide an automated technique for early detection of problematic batches, facilitating quick decision-making on potential batch recalls and improving overall quality management in manufacturing processes.

A wearable device has a body with one or more connectors for coupling the body to a lanyard. A camera assembly is mounted on the body. The camera assembly includes a pair of cameras configured to capture images of an environment surrounding the wearable device. The wearable device also includes a network adapter to transmit data derived from the captured images to a processing system.

An integrated artificial intelligence camera system provides multi-domain monitoring and enterprise automation from a single hardware endpoint. A camera assembly and auxiliary sensors, including at least one of depth, thermal, audio and environmental sensors, feed an edge compute module executing modular perception, identity, physiological estimation and domain-logic pipelines. The system detects and tracks persons, objects and activities; generates pseudonymous identity tokens for enrolled workers; estimates breathing rate and related wellness indicators from non-contact visual and thermal signals; and transforms these outputs into normalized event records annotated with security, inventory, workforce, payroll, physiological and compliance labels. A unified event model standardizes timestamps, device identifiers, site and zone identifiers, actor identifiers, metrics, confidence scores, policy tags and evidence references, enabling direct integration with payroll, inventory management, security incident and analytics platforms. A policy engine governs feature activation, anonymization and retention per jurisdiction and user role.

Systems and methods that apply generative artificial intelligence (AI) processes to facilitate the inventorying and moving of physical items. In certain embodiments crew members of a moving team are provided with mobile devices running an application to take an image of an item to be moved. Once the image is taken, the systems and methods initiate a generative artificial intelligence process to provide a human readable description of the item. This description and the image can be built into a file that is stored in an item database to create a visual catalog for a user. Room layout images can be captured and stored as well. The visual catalog provides a real time record of items being moved and the cartons holding the items. A portal into the visual database allows a client to deliver instructions where to locate or store an item being moved and the system can get those instructions to the crew in real-time.

A demand prediction display device includes: a memory configured to store instructions; and one or more processors configured to execute the instructions to: predict a demand of a consumer of a product in a prediction target period by using purchase record data of the product by the consumer; predict a demand for a wholesale shipment of the product in the prediction target period by using wholesale shipment record data of the product by a wholesaler; predict a demand for a manufacturer shipment in the prediction target period by using manufacturer shipment record data of the product by a manufacturer; and display a prediction result of the demand of the consumer, a prediction result of the demand for the wholesale shipment, and a prediction result of the demand for the manufacturer shipment on a display device on a common time axis.

The system may create marketing campaigns based on inventory data and/or sales data for a product. The system may compare the inventory data for the product or the sales data for the product to inventory threshold data or sales threshold data, respectively. The system may adjust a marketing campaign for the product based on a relationship between at least one of the inventory data and the inventory threshold data, or the sales data and the sales threshold data. The marketing campaigns may be further based on weather data, traffic data, event data, seasonal data, holiday data, calendar data, trend data and/or any other data.

A tire stock management apparatus (10) that calculates a stock quantity of tires during a specific time in future, the tire stock management apparatus including: a warehoused quantity acquiring unit (131) configured to acquire a predicted quantity of tires to be warehoused in connection with the specific time; an issued quantity acquiring unit (132) configured to acquire a predicted quantity of tires to be issued in connection with the specific time; and a stock quantity calculation unit (133) configured to calculate a stock quantity of tires at the specific time, based on the predicted quantity to be warehoused and the predicted quantity be issued, wherein the predicted quantity to be warehoused is calculated using warehouse data generated based on tire operation time data indicating a removal time, a usage time, and/or a delivery time of tires prior to the specific time.

An information processing system includes a memory and circuitry. The memory stores first information indicating a usage state of each of a plurality of devices in a storage facility based on information transmitted from the storage facility. The circuitry selects a device that is rentable from among the plurality of devices stored in the storage facility based on the first information stored in the memory. The circuitry transmits a selection result to the storage facility.