At PAHMA, we're planning for three large moves (whole rooms and whole buildings) of objects next year, and we want to manage the moves (including the pre-move inventories and containerization of objects, as well as the movement of objects to new normal/home locations). While the pre-move transition to barcodes will require much additional time and effort, our goal is to make this and future object movements (and inventories, spot checks, etc.) much quicker and easier, with our move- and inventory-related data being more complete and more accurate.
We want our preps, collections managers, and registrars to be able to digitally capture move-related data, but we want to minimize the risk to objects that inevitably comes with lugging additional gear into collections areas, having wires hanging/swinging near objects, and having one's hand(s) preoccupied with computers and barcode scanners.
We want to have a way to allow move- and inventory-related data to be gathered and stored quickly enough so as not to be an impediment to efficient inventorying, packing, and moving. Ideally, we'd interact directly with CollectionSpace, but if quick transactions (e.g., recording that an object is at a location) are not very rapid (ca. 1 second) when interacting directly with CollectionSpace, we'd need to save the data in a way that is very fast, and then do regular batch updates (at least nightly) to CollectionSpace.
We originally considered using lightweight tablets (iPads) that could be mounted on collections carts (and be removed when necessary), that would use a web-based interface supported by a wireless network, and that would be paired with a lightweight, battery-powered scanner 'fobs' (example1, example2, example3) that communicate wirelessly (Bluetooth) with the tablets. After discussing this with the Development team, several problems were highlighted: the reliance on wireless connectivity (if the network goes down, all work stops), reduced volume of work (work only progresses as fast as the web-based solution allows), and the expense of developing a web-based solution for the iPad.
As a result, we redirected our attention to network-independent bar code scanners with data storage capacity and a programmable interface. We looked at the many makes and models available and decided to test the AML M5900 (details). We purchased an M5900 with the optional 2D scanning capability and charging/docking station that allows easy up- and download of data.
Chris Hoffman used the Windows-based programming utility supplied with the M5900 to create a simple interface that allows a 2D location barcode to be scanned, followed by up to 100 1D object barcodes, with as many additional location (and corresponding objects) scanned as desired. The data is stored in an orderly CSV format, along with a timestamp for each scan.
Storage locations would be encoded with a Data Matrix 2D symbology, and objects would be bar coded with a Code 128 1D symbology, each with included text. Examples are shown below:
NB: until updated, the ?1D symbol above is actually code 39, not code 128.
After collecting data with the M5900, this data will need to be loaded onto a computer to begin its journey into CollectionSpace. These are the steps:
The scanner was dropping data if the user scanned too quickly. The handheld unit signalled apparent success with a beep, but we later discovered that objects which were scanned too quickly (measuring time since previous scan) never got saved to the .CSV data file. The problem was reported to AML, and AML technician Rick Shimmel revised the PAHMA Locations.ap5 program file to hopefully be less pace-limiting. To test this, we set up an experiment to scan objects in at several regular frequencies (measured in scans per minute): 40, 60, 90, 120/minute. These are the results:
Rate |
Old program |
New program |
Built-in P1 program |
---|---|---|---|
40/minute |
19/30 (37% missed) |
30/30 (0% missed) |
30/30 (0% missed) |
60/minute |
17/30 (43% missed) |
20/30 (33% missed) |
30/30 (0% missed) |
90/minute |
11/30 (63% missed) |
15/30 (50% missed) |
30/30 (0% missed) |
120/minute |
9/30 (70% missed) |
14/30 (47% missed) |
29/30 (3% missed) |