• 沒有找到結果。

4. Empirical Case Studies

4.2 Case: IT help desk service

4.2.1 The background of the case

The case study is a practical project that we had been involved in improving the service level of the IT help desk service division in a multinational company with its headquarters in Taiwan. The primary functions of the division cover the areas of the management of the company-wide computer network systems, supporting office automation activities, data processing and management, and IT technical consulting.

However, too often the divisional managers receive complaints from their employees and external customers that the IT service processing time is frequently much longer than they expected. Such an issue of lengthy service processing time has the potential to directly impact the employee’s work efficiency and the effective communications with the customer, and thus eventually affects the corporate performance and customer relationships. Other issues include a lack of a complete and sound Standard Operations Procedure (SOP) to guide the personnel in the division for handling the service requests in an efficient manner, insufficient number of employees to take care of the situation of the ever-increasingly crowded service requests, and the ineffectiveness of coping with the fast changing types of service issues confronting the existing outmoded hardware, software and other equipment.

4.2.2 Implementing the Lean Six Sigma methodology

Phase 1: Define/identify value

Step 1: The voices of both customers and employees can be identified as follows.

y Shorter IT service processing time needed to take care of service requests. This issue relates to the responsive dimension of the

service quality.

y Establish a SOP in order to efficiently respond to service requests by the employee and customer. This issue relates to the assurance dimension.

y An appropriate staffing plan for relieving the work overload on the employee. This issue relates to the responsive dimension.

y Replace or upgrade the current aging equipment to effectively handle any type of service issues. This is concerned with the

tangibles dimension.

Step 2: Concerning the key attributes identified in Step 1, we translate each of

those attributes into a measurable item as follows.

y A cycle time can directly link to the service processing time needed to take care of a service request.

y A process cycle efficiency relates to the amount of value-added time to the total lead time of the process (George, 2003). It can be a metric for measuring the work efficiency of the employee in the division.

y A throughput rate can be an indicator for examining the appropriateness of the staffing plan.

y A throughput index may reflect the effectiveness of the equipment used to handle any potential service issues.

Step 3: By applying the Kano model to identifying the CTQs, the metrics defined

in step 2 can be further analyzed based on the perspectives of the majority of 120 employees and 72 customers respectively who were randomly selected for answering the pre-designed questionnaires, and the results

were summarized in Table 1. Although the four metrics are all regarded to be important by either the employee or customer, however, since the cycle time has a direct impact on the other three metrics, therefore, it dominates over the others for the determination of the CTQs.

Table 4.8 The Kano model analysis for the determination of the CTQs

Metrics Customer’s perspective Employee’s perspective

Cycle time Performance Indifferent

Process cycle

efficiency Basic Basic

Throughput rate Indifferent Basic

Throughput Basic Basic

Phase 2: Measure/value stream mapping

Step 1: Once the CTQ, i.e. cycle time, was identified in the Define phase, a data

collection plan was developed to gather sample data. In this sample, 110 service requests in a total of 22,413 were randomly selected for a survey for a certain twelve-month time period before the project started. The data collected is summarized in Table 4.9.

Table 4.9 Data collected for the cycle time of the IT help desk service

CTQ

Average

(minutes/request) Std. deviation (minutes)

Sigma level (USL=180 minutes)

Cycle time 167.5 81.6 0.15

Step 2: Construct a current-state value stream map as shown in Figure 4.6. By

examining this map, one non-value-added step is identified and should be eliminated. It is the step which asks the user to provide correct and

* The upper specification level (USL) will be determined in the step 5 of the Measure phase.

complete information in order to expedite the service delivery.

Follow the SOP to handle the task assignment

for the service request

Inform the user of completion of the service

request

Figure 4.6 A current-state value stream map for the IT help desk service

Step 3: To construct a future-state value stream map as shown in Figure 4.7,

eliminate the non-value-added step from the process, and try to find any potential for reducing the cycle time in each of the remaining steps.

No

Follow the SOP to handle the task assignment

for the service request

Inform the user of completion

of the service request

Figure 4.7 A future-state value stream map for the IT help desk service

Step 4: Then, a detailed process map was developed as shown in Table 4.10. The

input/output items in each step of the whole process are identified and established whether or not they are controllable or uncontrollable items.

Table 4.10 A detailed process map for the IT help desk service

Process step VA /

VA IT help desk guideline list I Based on the procedure descriptions

C Check if the service

request is covered by the IT help desk service

VA Conclusion on whether the service request is covered by the IT help desk service or not

O N/A C

VA Check list for the service request

I N/A C

VA Level of the technical expertise possessed by the staff in the IT help desk service division

I L: at least one year of field experience

C Classify the service

request and assign it for processing

VA Task assignment for the service request

O N/A C

VA Type of the service request I N/A C

VA The impact level of the service

issue I N/A U

VA The position level of the user I L: divisional managers U Judge whether the

service request is urgent or not

VA Decision on whether to take action immediately or not

O N/A U

VA Request form I N/A C

VA Current job scheduling for an individual

I N/A C

Put the task assignment into the job scheduling

VA Waiting time for processing O N/A C

VA Problem confirmation I N/A U

VA Confirmation of the abnormal messages

I U: 3 minutes C

VA Confirmation of the user’s

computer environment I U: 3 minutes C

VA Confirm the authority levels for execution

I U: 3 minutes C

VA Confirm the operational

procedures I U: 3 minutes C

Check whether the information provided by the user is correct and complete

VA Confirmation of the correctness and completeness of the information provided by the user

O N/A C

Table 4.10 A detailed process map for the IT help desk service (continued)

NVA E-mail or call the user to provide

additional information I N/A C

Ask the user to provide correct and complete information

NVA Waiting time for the user to provide additional information

O U: 30 minutes C

VA Prepare a guideline for taking care of the abnormalities

.

I N/A C

VA Prepare for tools needed (hardware and software)

I N/A C

VA Coordination of related people I N/A U

VA Trouble shooting O U: 90 minutes C

VA Write a task report O U: 15 minutes C

Follow the SOP to handle the task assignment for the service request

VA Total processing time need for this

step O U: 120 minutes C

V/A Inform the user of completion of the service request by e-mail or phone

I N/A C

Inform the user of completion of the service request

V/A Time needed for receiving user’s response

O U: 30 minutes U

The user spends time on the confirmation

I U: 15 minutes C

Confirm with the user that the service request was resolved

V/A Result of the confirmation O N/A U

V/A The documents prepared for filing I N/A C

V/A Update the guideline for handling the abnormalities

O Update after the confirmation

C Document filing

V/A Close the job scheduling for the

service request O Update after the

confirmation C

Step 5: Referring to the data of the current specification levels for the cycle time

in each process step as shown in Table 3, we set the upper specification level for the total service processing time at 180 minutes, while the lower specification level is left blank since the service processing time is less, the better the service quality is.

Phase 3: Analyze/determine root causation

Step 1: By looking into the average cycle time data in each step of the

current-state and future-state value stream maps, it was easily discovered that there exist two major gaps between the ideal and existing amount of time spent on following the SOP to handle the task assignment for a service request and asking the user to provide correct and complete information respectively.

Step 2: The root causes of the problem of long service processing time were

identified by using the cause-and-effect diagram as shown in Figure 4.8.

Step 3: Among the root causes as depicted in Figure 4.8, the significant ones

were further identified by utilizing the Main Effect Plots as shown in Figure 4.9. We concluded that the root causes of the lack of a well-proven SOP to be followed for performing the task assignment and the inappropriate level of technical expertise possessed by the assigned IT staff would have a significant impact on the service processing time.

Figure 4.8 The cause-and-effect diagram for the problem of long service processing

time of the IT help desk service

Figure 4.9 Main effect plots for identifying the significant root causes of the IT help

desk service Long time period to

wait for user’s response to the notice of job completion

Lack of a solid SOP to be followed for performing the task assignment of the problem definition

Inappropriate level of technical expertise possessed by the assigned IT staff

Long time period to handle the task assignment for t Long waiting time for

user to provide correct and complete information

Personality issue

The working attitudes of the staff in the IT help desk service division

Misconception of the level of technical expertise required for the assignment situation of the I help desk serv

Degrees of appropriateness of the staffing plan Years of previous

working experience as an IT help desk staff Level of soundness of

the Standard Operations Procedure (SOP)

Rating Years Percentage

Service

Phase 4: Improve/flow and pull

Step 1: Based on the findings of the significant causes in the Analyze phase, the

solutions for eliminating these causes are proposed as follows.

y Examine and revise, if necessary, the current SOP in order to help deliver the services more smoothly and quickly. In addition, building up a knowledge base for the IT help desk service can also substantially benefit the shortening of the service processing time.

y Set a minimum requirement of at least 2-year previous experience as an IT help desk staff, and confirm his/her levels of technical expertise by the proven success record on the job assignment.

Step 2: A flow and pull system can be developed by following the process as

shown in the future-state value stream map. This new system features a streamlined process flow and specified customer-driven demand.

Phase 5: Control/pursue perfection

Step 1: Develop a control plan which involves the use of a Check List for the

internal strategic management, and an R-chart for the long-term service processing time control.

Step 2: As a control plan is established, implemented and completed, then, it is

recycled through the entire service process.

4.2.3 The implementation results and discussions

A follow-up study was conducted after the IT help desk service division had implemented the Lean Six Sigma methodology. A total of 437 sample cases were

randomly selected for this three-month study. The comparisons of results are summarized in Tables 4.11 through 4.1 as follows.

Table 4.11 Overall performance of the Lean Six Sigma project for the IT help desk

service

Before Target After CTQ

Average Std. dev. Average Std. dev. Average Std. dev.

Cycle time

(minutes) 167.5 81.6 112.0 49.3. 88.0 23.2

Table 4.12 Breakdown analysis by steps for the process of the IT help desk service

Average cycle time (minutes) Steps

Before Target After Check if the service request is

covered by the IT help desk service

2.5 1.5 1.5

Classify the service request and

assign it for processing 7.0 5.0 4.0

Judge whether the service request

is urgent or not 1.5 1.0 1.0

Put the task assignment into the

job scheduling 5.0 3.5 2.5

Wait for batch processing 45.0 30.0 22.0

Check whether the information provided by the user is correct and complete

6.0 5.0 3.5

Ask the user to provide correct

and complete information 25.0 9.5 8.0

Follow the SOP to handle the task

assignment for the service request 52.0 40.0 36.0 Inform the user of completion of

the service request 22.0 15.0 13.0

Confirm with the user that the

service request was resolved 8.0 5.0 5.0

Document filing 8.0 6.0 4.5

Table 4.13 The financial results of the Lean Six Sigma project for the IT help desk

service

Savings (US$/year) Target Actual

Hard savings 75,684 92,963

Soft savings 24,755 28,340

Total savings 100,439 121,303

One of the major findings in this study was that the average service processing time for a case of IT help desk service was reduced by 79.5 minutes or 47.5%

equivalent compared with the performance before implementing the Lean Six Sigma methodology, and it also exceeds the firm’s target number of 112 minutes. This achievement in cycle time reduction implies that the Lean Six Sigma methodology has proven its power to accelerate the process flow by eliminating the non-value-added steps while also streamlining the remaining value-added steps. In addition, it is encouraging to see that the standard deviation has decreased by 58.4 minutes or 71.6% equivalent. These results indicate that a more stable and consistent service processing time interval was established.

Using Six Sigma project terminology, this project achieved a 3.97-sigma quality level result combining both hard and soft savings valued at US$121,303 annually based on the calculations of the obtained data. The hard savings refer to the Six Sigma project benefits that allowed this firm to do same amount of business with fewer employees or handle more business without adding people. On the other hand, soft savings are Six Sigma project benefits such as reduced time to market, cost avoidance, improved employee morale, and other intangibles which contributed additional savings to the firm.