Service Process Performance Management

1
Service Process Performance Management

• Robert D. Weaver

r2w_at_psu.edu
2
Processes, Performance, Incentives

• The setting and the implications for service
  management
• Examples
• Current research potential for applications
• Example application projects
• Bottomline the potential

3
The setting

• Services provision involve complex processes
• Not typically linear
• Not typically based on fixed production rules
• Performance results from
• Human agent interaction
• Human agent management of resources, other
  agents, clients
• Achievement of performance requires
• Metrics to quantify and monitor
• Incentives direct agents

4
The issue How to manage performance?

• Management of service processes is fundamentally
  different from manufacturing processes
• Service processes
• Involve agents
• Agents are multifunctional and interactive
• Agents are self-directive implying control is not
  direct
• Agents embody tacit knowledge that is critical
  to performance
• Process performance often depends on
  collaboration among agents

5
Service performance management

• Two steps
• Define, implement, and monitor performance
  metrics of the service system
• Overall system performance
• Agent performance
• Process performance
• Design, implement, and monitor incentives that
  drive agent performance to achieve system goals

6
Step 1 Service process performance

• Salient features of service processes
• Role of agents
• Metrics for performance

7
Service systems are not linear
Marketing Sales
Technical Service
Outbound Logistics
Production
8
Service process characteristics

• Processes are fuzzy
• Semi-autonomous human agents
• Interact
• Loosely follow rules that describe process ideals
• Apply material and time inputs
• Respond to continuum of problem settings
• Agents manage subprocesses that compose a
  service system to deliver value

9
Service system configurations Simple cases
10
Service system configurations
11
Service system configurations
12
Service system configurations
13
Role of agents in service system

• Many agents
• Agent have decision-making capacity
• Agent goals affect decisions
• Agent manage sub processes
• Agents interact and are interdependent
• Monitoring is costly and imperfect

14
Characteristics of agents within a service system

• Directed by process rules
• Capable of decision-making in response to change
  in problem setting
• Guided by personal goals and resources that may
  include service system goals (alignment)
• Influenced by incentives

15
Types of agent interaction

• Asset service sharing
• Tactical integration joint modification of
  traditional silo operations to enhance
  collaborative performance
• Strategic integration

16
Multi-agent service system
17
Example Major League Baseball
18
What metrics for service performance?

• Metrics are simply indicators of performance
• Three types needed
• Agent performance
• Process performance
• Service system performance
• Performance must be defined by service system
  strategic goals and subsidiary objectives
• Quantifiable measures of necessary conditions for
  performance levels acheivement.

19
Prerequisites for metrics

• Observable
• Verifiable
• Low cost/value
• High frequency
• Leading indicators of service system goals

20
Examples of metrics

• Efficiency of resource use
• Fill rate, response delay, client revisions
• Financials typicals, cash-to-cash turn rates,

21
Example Major League Baseball
22
Why performance metrics?

• Monitoring for management
• Gauging net benefits of agent configuration
• Sharing net benefits through incentives
• Monitoring for diagnostics
• Identify opportunities for improvement
• Identify bottlenecks, new problems
• Establish opportunities for response
• Metrics form basis for incentive design

23
Step 2 Incentive design

• Align agent service system goals
• Manage agent performance

24
Alignment of goals

• Incentives signal service system goals
• Incentives encourage alignment
• Salary level
• Promotions
• Bonuses (cash, options)

25
Need for managing agent performance

• Agent autonomy creates bases for incentive design
• Agent performance may be difficult monitor on a
  continuous basis
• Agent potential (quality) difficult to measure
• Agent performance varies across agents
• Ultimately, agent performance reflects some
  voluntary human behavior !

26
Potential of incentives

• Differential incentives encourage productivity
  where pay-off will be greatest
• Incentive package can induce agent to
• Reveal true quality
• Perform at potential
• Incentive package design can align agent with
  service system goals
• Incentive package design can be done at the
  system level

27
Nature of the beast

• Choose incentives conditional on metrics
• Dynamic, stochastic optimization of service
  system goals, given
• Service system process
• Agent autonomy and optimization

28
Potential pay-offs

• Uniform incentives across heterogeneous agents
  fail to induce goal achievement
• Differential incentive packages provide means of
  inducing agents to contribute to system
  performance

29
Service system process management

• Summary

30
Three stages of research

• Service system mapping
• Characterize system processes, agents, and flows
  of effort that lead to delivery of service
• Performance metric design
• Derive metrics from service system strategic
  goals and system structure
• Incentive design