Multi-Skill Arithmetics: Insights and Challenges

undefined
 
April 16, 2024
Ger Koole, PhD
 
Webinar
Multi-skill arithmetics
Goal of this webinar
 
Insight in advantages & complexities of multi-skill operations
Rules of thumb of how to make multi-skill calculations
 
But first: who are you?
Poll: do you have multi-skilled operations?
Poll: which method do you use?
What are multi-skilled
operations?
 
Multi-skill: trade-off between
nobody & everybody multi-skilled
Focus on skill-based routing
Harder to manage and to plan
optimally
 
Everybody all skills is too
expensive to train/impossible
 
Everybody a single skill is not
flexible & not efficient
What makes multi-skill planning
difficult?
 
Single skill: 1 number per interval
N skills: potentially 2
N
-1 skill groups and thus 2
N
-1 numbers
For example, 3 skills: sales regular (R), sales premium (P) , admin (A)
Possible skill groups: R, P, A, RP, RA, PA, RPA
Often not all combinations possible: e.g., R, RP, A, RA, RPA
 
No simple way to predict performance
No formula as for single-skill case (Erlang)
Only (crude) approximations
Simulations only accurate but slow option
Result: optimization difficult
The objective
 
Upper and lower bound:
You can never do better than fully multi-skilled
You can never do worse than fully single-skilled
But: multi-skilled agents cost more
Trade-off between ”cheap” single-skilled agents and “efficient” multi-
skilled agents
 
multi-skill % →
 
agents hours
 
av. costs
per agent hour
 
total costs
multi-skill % →
multi-skill % →
How much “multi-skilledness”?
 
20% is based on extensive
experiments, but can be
different depending on
situation
Curve is flat thus % does not
need to be very precise
 
multi-skill % →
 
total costs
Example
 
FC=5/min
 
FC=4/min
 
AHT=3min
 
AHT=5min
 
patience=5min
 
patience=5min
 
18
 
23
 
84%/20s
 
81%/20s
 
39
 
85%/20s
 
85%/20s
 
13
 
18
 
8
 
80%/20s
 
82%/20s
 
AHT=3min
Robustness
 
Forecasts are always wrong – and sometimes really wrong
 
 
 
 
 
 
 
How can we make the SL even more balanced?
Better routing
More multi-skilled agents
 
 
 
 
 
 
 
 
 
 
FC=4.5/min
 
FC=4.4/min
 
AHT=5min
 
18
 
23
 
92%/20s
 
67%/20s
 
13
 
18
 
8
 
83%/20s
 
75%/20s
The role of routing and intra-day
management
 
Bad routing can make the situation worse than single-skill
Routing is most of the time “static”
Not adapted to the current situation
Requires ”real-time” management (RTM)
Makes the situation worse…
 
 
 
 
13
 
18
 
8
 
80%/20s
 
78%/20s
 
Multi-skilled agents: priority to 2
 
13
 
20
 
6
 
80%/20s
 
77%/20s
 
RTM: more agents for skill 2
Evaluation
 
SL of SBR is always lower than fully
multi-skilled 
 add 1 agent
Assuming AHTs stay the same
For robustness (errors) add 10% more
 
Shrinkage should be added
Long-term: additional flexibility for capacity fluctuations
>2 skills:
Make sure all skills are “connected”
Better many agents with few skills than few with many
 
 
 
 
 
 
 
 
 
 
12
 
16
 
12
 
87%/20s
 
87%/20s
Interested in CCmath tooling?
 
Interested in a demo of one of our products?
Contact Jesper at
 
jesper@ccmath.com
 
Advice on how to use one of our tools or other WFM questions?
Book a free 1-hour slot with Bas and/or Ger
Contact us at 
bas@ccmath.com 
or 
ger@ccmath.com
 
Next webinar: 
May 28
Thanks for your attention!
Slide Note
Embed
Share

Delve into the intricacies of multi-skill operations in the webinar presented by Ger Koole, PhD. Explore the advantages and complexities, learn rules of thumb for multi-skill calculations, and understand the trade-offs involved in managing multi-skilled teams efficiently.

  • Multi-Skill Operations
  • Webinar
  • Arithmetics
  • Ger Koole
  • Challenges

Uploaded on Sep 18, 2024 | 1 Views


Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.

E N D

Presentation Transcript


  1. Webinar Multi-skill arithmetics April 16, 2024 Ger Koole, PhD

  2. Goal of this webinar Insight in advantages & complexities of multi-skill operations Rules of thumb of how to make multi-skill calculations But first: who are you? Poll: do you have multi-skilled operations? Poll: which method do you use?

  3. What are multi-skilled operations? Everybody all skills is too expensive to train/impossible Everybody a single skill is not flexible & not efficient Multi-skill: trade-off between nobody & everybody multi-skilled Focus on skill-based routing Harder to manage and to plan optimally

  4. What makes multi-skill planning difficult? Single skill: 1 number per interval N skills: potentially 2N-1 skill groups and thus 2N-1 numbers For example, 3 skills: sales regular (R), sales premium (P) , admin (A) Possible skill groups: R, P, A, RP, RA, PA, RPA Often not all combinations possible: e.g., R, RP, A, RA, RPA No simple way to predict performance No formula as for single-skill case (Erlang) Only (crude) approximations Simulations only accurate but slow option Result: optimization difficult

  5. The objective Upper and lower bound: You can never do better than fully multi-skilled You can never do worse than fully single-skilled But: multi-skilled agents cost more Trade-off between cheap single-skilled agents and efficient multi- skilled agents per agent hour agents hours av. costs total costs multi-skill % multi-skill % multi-skill %

  6. How much multi-skilledness? 20% is based on extensive experiments, but can be different depending on situation Curve is flat thus % does not need to be very precise total costs multi-skill %

  7. Example patience=5min FC=5/min 18 84%/20s 85%/20s AHT=3min 39 FC=4/min AHT=5min 23 81%/20s 85%/20s patience=5min 13 80%/20s 8 82%/20s 18

  8. Robustness Forecasts are always wrong and sometimes really wrong FC=4.5/min 13 83%/20s 18 92%/20s AHT=3min FC=4.4/min AHT=5min 8 75%/20s 23 18 67%/20s How can we make the SL even more balanced? Better routing More multi-skilled agents

  9. The role of routing and intra-day management Bad routing can make the situation worse than single-skill Routing is most of the time static Not adapted to the current situation Requires real-time management (RTM) Makes the situation worse RTM: more agents for skill 2 Multi-skilled agents: priority to 2 13 13 80%/20s 80%/20s 6 8 77%/20s 20 78%/20s 18

  10. Evaluation SL of SBR is always lower than fully multi-skilled add 1 agent Assuming AHTs stay the same For robustness (errors) add 10% more 12 87%/20s 12 87%/20s 16 Shrinkage should be added Long-term: additional flexibility for capacity fluctuations >2 skills: Make sure all skills are connected Better many agents with few skills than few with many

  11. Interested in CCmath tooling? Interested in a demo of one of our products? Contact Jesper at jesper@ccmath.com Advice on how to use one of our tools or other WFM questions? Book a free 1-hour slot with Bas and/or Ger Contact us at bas@ccmath.com or ger@ccmath.com Next webinar: May 28 Thanks for your attention!

Related


More Related Content

giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#