How do we choose the RFID asset platform? Functional coverage and extension capacity assessment
RFID Knowledge

How do we choose the RFID asset platform? Functional coverage and extension capacity assessment

PublishedShanghai Zejie Smart Technology Co., Ltd.

Picking up the RFID asset platform, most people are stuck between "Full but expensive" and "Count-based." The criteria are two: whether the core scene runs, or expands in three years or not. See if the base is all on the inventory, position, alerts that the interface and secondary development can match existing ERP and financial systems. The functionality covers today, and expands capacity to determine whether it will work for the next three years.

Functional coverage assessment

The functional coverage is not looking at how many items are on the list, but at what depth your business really relates to. Before you pick a platform, you draw the core process of your asset management -- from the procurement library, from the transfer, from the inventory, from the inventory to the end-of-life, and from each step to the platform ' s functionality.

The inventory function is fundamental. A plant that does warehousing, the most important concern is whether handheld terminals or fixed readers can quickly identify large-volume labels, rather than read them one by one. Batch reading rates (typically 50 to 200 tags per second in industry), drop rates (1% to 3% typical level), and resistance (metal, recognition distance decay in liquid settings) are the focus of assessment.

The location function varies considerably. The RFID itself is of limited precision, room-level positioning (in error 3-5 m) is low and easy to achieve; regional positioning (in error 10-30 m) requires additional emulators or reference labels, and input increases significantly. If the business needs only "know which floors the asset" it will be enough to select a room level; if you are to track down a particular shelf or cabinet, you need to evaluate the accuracy of the position and hardware input in advance.

The fact that the platform supports a self-defined field for an asset card, changes in records, volume imports and exports, aligns with the financial depreciation cycle, is very much influenced by the later use of details. The same key to authority management -- unusual movements of assets, unauthorized access to the bank, uncounted over time, etc. -- is self-defined, and the ability to subdivided rights by department, role, asset class directly affects the management of particle size.

Extension capacity assessment

The expansion of capacity determines whether the platform can follow business. The assessment focuses on the level of openness to interfaces, the degree of freedom to develop secondaryly and the cost of docking with existing systems.

The quality of the interface document is the first threshold. A complete API document should contain a list of interfaces, examples of requests and responses, error code descriptions, call frequency limits. If the manufacturer only gives SDK no documents or documents are not updated for long periods, subsequent integration will be very tiring. The vendor can be asked to provide an unsensitized API example, see if field names are regulated, and whether the error process is clear.

The secondary development of capacity is at two levels: whether the platform supports a custom workflow, a custom report, a custom field; and whether the bottom data model is open, whether the developers can read and write business sheets directly or only through limited API operations. The more "speech" the platform is open, the lower the cost of customization later.

The cost of docking with existing systems is often underestimated. A RFID asset platform is actually used, almost always with ERP, the financial system, OA. There are three ways to connect: a database that is directly connected (high risk, careful use), an intermediate table synchronized (flexible, recommended), an API in real time (cleanest but developed). Each approach has a very different implementation cycle and human input, which is clear enough to make less curves before and after.

Shanghai Select Technologies, Inc. found in past customer integration projects that the selection phase was focused on functional lists, ignored extension capabilities, and often found that later on when you wanted to take over an ERP, interfaces were restricted, but only system-replaced or expensive. So at the selection stage, the weight of the expansion capacity should be equal to the functional coverage, even higher.

Cost structure dismantled

The selection platform cannot be based solely on software offers, but it is necessary to separate the obvious costs from hidden costs.

Visible costs include software (in terms of number of users, number of assets or a charge for the same amount), implementation service charges, annual maintenance fees (usually 15-20% of license), hardware purchases (readers, labels, gateways). Based on the size of 5,000 asset points, initial input of software and hardware is typically in between 300,000-800,000 entangles, depending on functional modules and brand location.

The hidden costs are easily overlooked: internal staff training (usually 1-2 weeks), data migration and cleansing, parallel operation of systems for transition, and the development of a third-party system. Together, hidden costs may account for 30 to 40 per cent of total inputs, and these money spends no visible output.

The costing model is also clear. User-to-user rates are appropriate for small teams, asset-to-asset costs are suitable for stable size, and modular charges are appropriate for the gradual expansion of demand. If the size of future assets doubles, programme costs on asset counts will increase very quickly, and it is recommended that they be measured in advance.

Apply scenes to scale

The needs of enterprises of different sizes and industries vary widely, and the platform type cannot replicate others ' lists.

Small and medium-sized enterprises (with 500-5000 assets) give priority to lightweight SaaS programmes, which are fast deployed, low initial inputs, and with a functional iterative approach that is handled by the manufacturer. These types of programmes usually have limited capacity to expand, but will be sufficient for three or five years if business stability prevails.

Large groups (more than 10000 assets) require private deployment to support multi-organizational, multi-authority and in-depth customization. Such programmes are initially well invested, with long implementation cycles (usually 3-6 months), but expanded capacity and data autonomy.

The difference is also evident in the industry. One hospital that manages medical equipment, most concerned with tracking and inventory accuracy of high-value consumables; one company that leases IT equipment, which values the automation of access to asset banks and client billing. These two types of business require a completely different platform, depending on whether the manufacturer has any actual case in the same sector.

Common pits in selection and implementation

The function is very beautiful, and it's completely different. The manufacturer's presentation usually uses well-prepared data, size, and anomalies are covered up. It suggests that the manufacturer should pilot a process in your environment to run a real data flow to see if it meets daily needs.

The unwritten "free upgrade" in the contract is a fee item when it is implemented.

The platform is better and bottom hardware is not readable. The readings are tested in the application environment, especially for metal, liquid, high-density stacks.

The training and documentation support is not in place. Once the platform is online, no one inside will answer questions, which are more lethal than a functional failure.

Selecting RFID asset platforms does not have a standard answer. Evaluate functional coverage and extension capabilities separately, first to do small-scale pilot testing, then to do long-term costing, much more stable than a board. The function determines whether today it will be available, if it will be regretted in three years -- in that order, the selection decision is much clearer.