RFID Observations and tag detections with zones and coordinates on a CAD floorplan

This CAD_Design_RFID_Data.csv dataset connects RFID tag detections with zones and coordinates on a CAD floorplan. In a supply chain, this structure can support tracking tagged products, pallets, containers, or equipment within a facility. The file does not identify which physical asset each tag represents. The uploaded file contains 956,816 rows and 13 columns, covering 308 distinct nonmissing tag IDs, 3 reader identifiers, and 12 zones on a single floorplan. What each column means Column Meaning in this dataset Supply chain use TagUId RFID tag identifier Identifies a tagged asset once linked to an inventory record. TagScannedId Contains three IP addresses, apparently identifying readers Indicates which reader detected the tag. TagName Contains “Unregistered” or “0000” Intended descriptive field. It currently provides little asset information. TagAntenna Antenna index, from 0 to 3 Identifies the reader antenna associated with a detection. TagCount Recorded detection count Can help characterize how frequently a tag is observed. Its counting and reset rules need confirmation. TagFirstDiscovered First detection time field Potential starting point for a visibility or dwell interval. TagLastSeen Most recent detection time field Potential endpoint for tracking visibility. ZoneUId Unique zone identifier Links the observation to a defined facility area. ContainerId Labels from LabZoneA through LabZoneL Functions as a zone label here. The values do not establish that these are physical shipping containers. X Horizontal floorplan coordinate Supports spatial mapping. Y Vertical floorplan coordinate Supports spatial mapping. FloorplanUId Floorplan identifier Associates the record with the facility layout. RSSI_dBm Received signal strength, ranging from −85 to −30 dBm Supports analysis of signal patterns and possible location estimation. How this relates to a supply chain A typical workflow would associate each TagUId with a business asset, such as a pallet of parts. Reader detections would then help establish which facility zone contains that asset. For example, if LabZoneA represented receiving, LabZoneF represented storage, and LabZoneL represented shipping, validated changes between these zones could support: Inventory visibility: Determine which tagged assets are present in each area. Movement tracking: Identify transfers between receiving, storage, production, and shipping. Dwell analysis: Measure how long assets remain in a zone once complete timestamps are available. Exception detection: Flag assets observed in unexpected zones or no longer detected. Spatial analysis: Display observations on the CAD layout and investigate signal coverage. These are potential applications. The file’s actual zones are named LabZoneA through LabZoneL, and their business functions are not specified. What an example record tells you The first record associates tag E280689000000000853D3037 with reader 192.168.1.42, antenna 2, and LabZoneC. It records coordinates (273, 350) and signal strength −82 dBm. That means the dataset associates this tag observation with that zone and coordinate. Confirming whether the coordinate represents a measured tag position, an assigned reference point, or another CAD feature is necessary before interpreting it as the asset’s physical location. Important findings before using it for research or operations Repeated observations need investigation. Many records share the same tag, reader, antenna, count, time fields, and zone while differing in coordinates and RSSI. Grouping by those shared fields produces 231,093 groups, with at least four rows in every group. This pattern could reflect augmentation, multiple spatial reference points, or another expansion process. The CSV alone cannot establish which. Consequently, its 956,816 rows should not automatically be described as independent physical measurements. The timestamps are incomplete. Values such as 34:16.7 lack a date and hour. They cannot reliably establish event order, elapsed time, or supply chain dwell duration without their original format and context. Coordinates need a scale and definition. X ranges from 18 to 703, and Y ranges from 134 to 600. Their physical units are unspecified, so these values cannot yet support location errors stated in meters. Asset information needs enrichment. The tag names are placeholders. Linking tag IDs to SKU, pallet, shipment, order, or equipment records would turn detections into useful inventory information. Data quality and model validation require care. There are 16 missing tag IDs and 430 exact duplicate rows. Related or expanded records should stay together when splitting training and test data to avoid overstating model performance. Also, because ContainerId maps directly to ZoneUId here, using ContainerId to predict ZoneUId would reveal the answer to the model. For a RFID spatial research, this dataset provides a useful connection between tag identity, reader observations, signal strength, and a CAD layout. Establishing the origin of the repeated rows, the meaning of X and Y, and the complete timestamps will determine how strongly it supports claims about localization accuracy and supply chain movement.