Back by popular request: Thing to keep in mind as you deploy mission-critical Auto-ID applications

source: www.istockphoto.com

In recent conversations with industry professionals, a recurring theme that comes up is ‘How do I approach a Auto-ID based solution like any other enterprise application roll-out’? A simple enough question at first-glance, but as I prodded a bit more, it became clear that certain factors seemed to be causing apprehension in their minds – I’d like to share these insights gained for the benefit of the larger audience, and the questions behind the above question roughly translate to the following customer asks (with some paraphrasing by yours truly):

  • Any solution we deploy needs to be Auto-ID hardware agnostic i.e. Why can’t I chose the most effective form of data-capture (RFID, barcode, sensors etc.) based on my application requirements (vs. what’s supported by the vendors software), and this needs to be able to ‘scale out’ to multiple devices and device types over time.
  • The new solution must extend and ‘make better’ my existing Line-of-Business systems through Auto-ID data –  not try and replace them!
  • ‘Mission-criticalness’ is a must… if I am going to replace existing ad-hoc or manual operations with it, that means 24×7 ‘hot-hot’ availability and failover metrics similar to my application and database servers. And it needs to play well within my existing enterprise platform architecture (ala Microsoft or IBM or Orcale) vs. custom middleware or worse, custom code masquerading as a proprietary development kit on device / server.
  • Need packaged applications that address the common requirements across deployments like being able to specify a configurable asset catalog, support for asset commissioning, zone audits, check-in, check-out and movement detection etc.
  • Remote Management of devices must be through standard mechanisms and tools – can’t be a silo or require any custom software on the devices which only leads to additional manageability nightmares when something goes wrong in production
  • Mobile applications are an integral part of scanning / error-checking operations on the ground and need to work across various form factors such as hand-helds, fork-lifts and kiosks. We expect manageability support here to be standard including ‘over the air’ provisioning of device applications and support for sporadically connected / disconnected environments  with the central application server.
  • My IT team HATES the complexity associated with custom code – customization and configuration need to be accomplished with standard development tools / IDE’s, not proprietary tools

To address the above, I went back and dusted some of the earlier posts on this subject – and they still seem to hold water going by the feedback of our clients – so here there are once again for your perusal:

    The Top 5 factors that most influence a successful deployment of RFID applications in your Enterprise

    The Top 5 habits of all successful, packaged RFID applications

     

I hope you find these resources useful and educative. Please do not hesitate to e-mail or call me if you have any questions, or would like to discuss how our Spotlight asset tracking platform and industry solutions  address the above and might be useful for a scenario you are planning.

Cheers!

/a

Digg This
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s