What makes a scanner a “Production” scanner?

Document capture industry experts share a general consensus around the concept of “Production” level status as an elusive threshold that separates low to high volume document capture. “Volume” is usually measured in deliverable documents per month as opposed to images per minute.

A “Production Level” document capture workflow can be represented by a complex graph starting with the ingestion of source documents (paper, micrographic, born digital) and ending with approved deliverables. The nodes of the graph handle scanning, QA/QC, repairs, image cosmetics, image processing, classification, coding, indexing, formatting, validation, publishing, deployment, auditing, reporting and more. Scanning nodes may or may not end up as part of the critical path in that graph. In turn, a scanning node can be represented by a sub-graph that includes scanning equipment as a sub-node which again, may or may not be part of its critical path.

For the hardware component of a scanner to qualify as production level, it must meet or exceed certain thresholds based on:

· Throughput: measured in compliant images per minute. Compliant images meet service level agreement acceptance criteria, do not need rescan and are not mutilated, overlapped, duplicated or missing.

· Trustworthiness: Must be dependable, reliable, robust and durable. High volume throughputs need to rely on it.

· Versatility, resilience: Must smartly adapt to the many challenges of real world media and should not require frequent human attention during capture. It requires technological sophistication in cameras, light sources, feeding mechanisms, error detection and recovery, etc.

· Batch versus transactional: The unit of work in a production level capture workflow is a batch (collection of images) managed by combinations of human and automated processes. All production level workflows hinge on automation and batch processing. Stop and go, transactional, on-demand processes and the like do not belong to the production realm.

The software component of a scanner plays a very important role in accomplishing all of the above. However, a line must be drawn separating intrinsic from supplemental software functionalities. To fairly qualify a scanner as production level, only intrinsic, integral software functionalities count. Supplemental functionalities escape the scanner domain, as they belong to the production workflow instead.

Finally, we must include the cost-effectiveness component, which includes factors such as productivity, cost and risks. A scanner that is highly dependent on human labor or fails to satisfy the criteria described above is likely to squander the initial cost savings before the high volume project is completed. Also, the risks may outlive the project, as some of the fallouts of using a sub-production scanner may start popping up weeks or months after final project acceptance.

In conclusion, a scanner can be only considered a production scanner if when plugged into a high volume production workflow, meets or exceeds SLA volume thresholds while sustaining an acceptable level of cost-effectiveness.