I asked my Professor, “Prof., can we adopt the “Auto Scaling” feature in Work Force Management”. He said, “Yes, why Not!”
Well, this has been the Starting point of our journey. Let me first delve into ‘Auto scaling’
What’s Auto Scaling?This is one of the most popular features of any Cloud Platform. Going by the definition – Auto scaling adjusts the ‘resource capacity’ as needed. It is the answer to the most critical questions like
a) How can I ensure that we have enough resources catering to my fluctuating need?
b) How can I automate provisioning of resources to cater to our demand?
Based on the conditions (“Auto Scaling Policy”), resources will be added (i.e. Scaling Out) and resources will be taken out (i.e. Scaling In)
Now, How This Can be Adopted in Work Force Management?
A) The Format Of the Game : In an Agile environment, the tasks are divided into ‘Squads’ under different Squad leaders & its members. The Squads are formed based on certain parameters like Objects (& its Criticality) to be delivered wrt the Sprint Planning. Members (nos. & skills) are decided & grouped based on the said considerations.
B) Emergence Of Different Scenario :With festive seasons around, Client wants to push more (‘Offers & Discounts’) objects to the ‘production server’ and make it ‘Live’ for their Customers. It calls for increased ‘demand’ for the Work Force Management team
C) Auto Scaling Implementation : Auto Scaling can be implemented here. The Wfm_Watch (is a ‘service’) would trigger alarms and calls for initiating the Auto Scaling Service.
By Wfm_Watch, we mean the Metrics which are coming out of the Agile Systems of records. There would be a group of parameters (e.g. utilization percentage of members >90% in the last 5 working days) which would signal the Increased load on a particular Squad. The Metrics (Jira etc.) of the Agile system of records would indicate/ forecast the higher Utilization of the Members of the Squad
Auto Scaling Service is comprised of the Guild leaders. They would validate the ‘increased’ resource provisioning based on the ‘trigger’/’alarm’ from Wfm_Watch. The Auto Scaling Service would recommend ‘Scaling Out’ i.e. provisioning of additional resources (here members) to Squad
Tribe Leader, based on the above recommendation, would give the Direction to implement the same. The Tribe leader, being focused on value creation/growth and serving customers, would monitor the Cost & Utilization of the team member.
In continuation to that, once the Objects are delivered and (for the squad) it is coming to its normal level, the same ‘Auto Scaling Policy’ would take away the Team Members & ‘Scaling In’ happens.
What’s the Advantage?
a) Cost goes down: The capacity across the Squads remains balanced & thus the overall cost-to-implement is low
b) Quick To Implement: Since this is nothing but Internal allocation of Team Members (as against Hiring someone from outside), this would be quick & effective
c) Customer Delight & higher NPS: Importantly, the ‘new’ pool of resources will be equally skilled having the same domain expertise. This would result Customer Delight & higher NPS
I would like to have your perspective here.
Pls note that this is absolutely my thought-process and has got nothing to do my organization
Author : Sivaji Banerjee, WFM Programme Manager at IBM