Integration of Capability Analysis and Its Impact on Process  
Capability: Evidence from the Constituency Development Fund  
Process for Project Approval and Funding in Zambia  
1Kelvin Kashindi, 2Joe Likando, 3Francis Simui  
1PhD Candidate, University of Zambia and Zimbabwe Open University under the Institute of Distance  
Education  
2Dr Joe Likando, Post Graduate Supervisor, educationist, Public Administration specialist, University of  
Zambia  
3Professor, Francis Simui, Assistant Director University of Zambia, Institute of Distance Education,  
Researcher, Supervisor, Lecturer and Mentor  
Received: 27 November 2025; Accepted: 01 December 2025; Published: 09 December 2025  
ABSTRACT  
Processes by nature are affected by variation, either chronic or sporadic, which hinder them from consistently  
producing desirable outcomes relative to specifications. Existing research has rarely utilised analytical tools that  
quantify variation and capability within the constituency development fund process (CDF). They rely on  
qualitative or descriptive assessments, which do not capture the extent to which variability contributes to process  
delays. The study aimed at demonstrating how integrated capability metrics provide objective, quantifiable  
measures of process variation to capture the extent to which it contributes to process capability. For example,  
the use of descriptive statistics on selected projects revealed differences in mean, standard deviation and sigma  
levels that could not be captured by qualitative or descriptive assessment. The study used a quantitative-  
descriptive research design. Data were gotten from the Value Stream Map developed for CDF process for project  
approval and funding. Data analysis was done descriptively through Median, and Inter Quartile Range (IQR) to  
capture variation and its impact on process capability and funding. Key Result(s): while qualitative or descriptive  
assessment provide subjective impressions of process performance, it fails to capture the magnitude, distribution  
and implication of variability on process capability. The differences in median (14) and IQR (21), indicates high  
variability. Hence, the process is unstable and unpredictable, leading to delayed project implementation, under-  
utilization or late utilization of funds. Recommendation: There is need to prioritize and improve data collection  
and record keeping to stabilize the process and enable better forecasting and capability monitoring. Conclusion:  
Improving process stability should be the first step before meaningful capability analysis can be performed.  
Key Words: Variation, Process stability, Approval and funding Time, Cycle Time, Digital Tracking,  
INTRODUCTION  
The constituency development fund (CDF) is a public fund mechanism, established to accelerate community  
development in various constituencies. However, the utilization of the said fund has faced a lot of recurring  
challenges in Zambia, for example, underutilisation and bureaucratic delays in approval and funding, leading to  
compromised development. In order to address such challenges, it is critical to integrate capability analysis so  
as to capture the extent to which variation affects process capability a reality that can never be captured by  
qualitative or descriptive assessment only as most studies have demonstrated (Kwilinski and Kardas,  
2024).Quantity determines the measurable aspects that give structure to qualities; the manifestation of quality is  
inherently linked to its quantifiable aspects. Without quantity, quality remains an abstract concept, elusive and  
impractical. Therefore, by grounding the philosophical abstraction of quality in the quantitative reality of process  
capability indices, the gap between theory and practice is bridged (Kwilinski and Kardas, 2024).  
Page 982  
METHODOLOGY  
Research Design  
The study utilised a quantitative-descriptive research design to capture the extent of variation through descriptive  
statistics in the CDF process.  
Objective  
To demonstrate how integrated capability metrics provide objective, quantifiable measures of process variation  
Sampling Frame  
The sampling frame comprised all process steps in the CDF process that have cycle time namely, proposal  
development(1-2weeks), technical appraisal(1-2weeks), submission of recommended projects(1week),  
ministerial and treasury approval(2-4weeks), fund disbursement(1week) and procurement(1-3months).  
Sampling Technique  
The study utilised purposive sampling by targeting only process steps with cycle time as identified above.  
Data Source instruments  
Data was obtained from the developed value stream map of the CDF process and personal interviews  
Data Analysis and interpretation  
Data was analysed through descriptive statistics and interpreted by comparing the differences between the mean  
and standard deviation and range to determine the extent of variation.  
Capability Analysis  
Montgomery (2013), describes Capability analysis as a set of activities that quantify variability of a process,  
analyses it in relation to the requirements and specifications of the product and to support the reduction of the  
variability. Capability analysis finds its expression in capability indices or metrics (Yang, 2023, Swamy and  
Nagesh, 2014). The natural existence of any process is that it always has variability, either common cause or  
special cause (Namane, 2023). As such, understanding and managing it is critical to solving issues of inconsistent  
quality, delays, or rising costs (Peterka, 2024). It is of paramount importance to always measure processes in  
order to get a true picture of the size of variation. Without a true picture of process variation, it is impossible to  
come up with an appropriate action plan. Therefore, tools like standard deviation, mean and range, are metrics  
that identify, quantify and monitor variation. They are the primary gauges and dashboards on which action plans  
are based (Fernandes, 2022). It is therefore impossible to even consider continuous process improvement without  
capability analysis.  
Variation  
process variation refers to uncontrolled or unexpected differences in a process’s outputs (Six Sigma, 2024)  
Types of variation  
Fernandes (2022) postulates that within common cause variation, there exists within subgroup and overall  
variation. The former is critical for technical process improvement while the latter is what affects the  
stakeholders or beneficiaries whether they are satisfied or otherwise. It is important to note that while the two  
types of variation exist, in common cause variation, there exists within subgroup variation and overall variation.  
In the former, the standard deviation is useful for understanding the variation within groups, excluding the  
variation between groups, offering a more accurate representation of the inherent variability within each group.  
Page 983  
In the latter, the standard deviation provides a broader measure of variability across the entire dataset. Therefore,  
it is imperative to compare the within-subgroup standard deviation with the overall standard deviation because  
a substantial difference between them is a strong signal that the process may not be stable, or that the process  
has other sources of variation in addition to the variation within subgroups. Hence, the utilization of control  
charts is critical to verify process stability before performing capability analysis (Fernandes, 2022).  
In order to capture the extent to which overall variation affects process capability and funding in the CDF  
process, the study only computed the various process steps with cycle times as shown in figure 1 of the current  
CDF process because there was not enough data to compute withing subgroup variation.  
Figure 1: Current CDF Process Flowchart  
Figure 1 above shows that out of 13 process steps, only 6 have cycle time namely, Proposal development: 1-2  
weeks, technical appraisal: 1-2 weeks, Submission of recommended projects: 1 week, MLGRD and Treasury  
approval: 2-4 weeks, Fund disbursement: 1 week and procurement: 1-3 months. Table 1 shows a step-by-step  
lead time calculation derived from the developed Value Stream Map (VSM) of CDF. All durations were  
converted to days (1 week = 7 days, 1 month = 30 days).  
Table 1: Lead time components for CDF project approval and funding process (minimum, maximum and  
midpoint).  
Activity  
Minimum Maximum Midpoint (days) Midpoint  
(days)  
(days)  
14  
14  
7
(weeks)  
Proposal development by WDC  
Technical appraisal by DP & DES  
Submission of recommended projects to MLGRD  
Approval by MLGRD & Treasury  
Fund disbursement  
7
10.5  
10.5  
7.0  
1.5  
7
1.5  
7
1.0  
14  
7
28  
7
21.0  
7.0  
3.0  
1.0  
Procurement  
30  
90  
60.0  
8.57  
Page 984  
KEY RESULTS  
The key results were based on the values in table 1 above and was done in two parts, 1st calculation of the  
minimum total lead time (sum of specified minimums): 72 days (10.29 weeks, ~2.4 months), the maximum total  
lead time (sum of specified maximus): 160 days (22.86 weeks, ~5.33 months) and the average (Midpoint) total  
lead time: 116.0 days (16.57 weeks, ~3.87 months). The 2nd part was to analyse the absolute values in the  
Maximum number of days column using descriptive statistics. The first task was to ascertain the distribution of  
the data set. Measures of central tendency were calculated to determine which one should be used, mean and  
standard deviation or median and interquartile range.  
Based on the raw data set concerned (14,14,7, 28,7, 90), the data distribution was found to be not normally  
distributed hence the median and interquartile range were used to determine the extent to which variation affects  
process capability.  
Median = (X[n/2] + X [n/2 + 1]) / 2  
n = 6 (even number) Position 1: n/2 = 6/2 = 3rd value = 14 Position 2: n/2 + 1 = 4th value = 14. Median = (14 +  
14) / 2 = 28 / 2 = 14  
Range = Maximum – Minimum, therefore, Range = 90 - 7 = 83,  
Interquartile Range (IQR), Q1 (25th Percentile), Position = (n + 1) × 0.25 = (6 + 1) × 0.25 = 1.75, Since  
1.75 is between positions 1 and 2, interpolation was made: as such, Q1 = X[1] + 0.75 × (X[2] - X[1]) Q1  
= 7 + 0.75 × (7 - 7) Q1 = 7 + 0 = 7 and the last part was to Calculate Q3 (75th Percentile). Actually, for  
n=6, using Tukey's hinges method Q3 =28 since it is the median of upper half: (14,28,90) and by the  
same method confirm Q1 =7 since it is the median of lower half: (7,7,14).  
In short: Median = 14, Q1 = 7, Q3 = 28 and IQR = Q3 - Q1 = 21  
The determination of skewness was done using Bowley's Coefficient = (Q3 - 2×Median + Q1) / (Q3 - Q1).  
Therefore, the computation was as follows:  
(28 - 2×14 + 7) / (28 - 7)  
= (28 - 28 + 7) / 21  
= 7 / 21  
= 0.33 as the Bowley’s coefficient representing a moderate positive skew.  
The data set also revealed that the value 90 was the outlier by use of IQR method as follows;  
Lower fence = Q1 - 1.5 × IQR Upper fence = Q3 + 1.5 × IQR, by substitution, the computation was as follows:  
Lower fence = 7 - 1.5 × 21 = 7 - 31.5 = -24.5 and Upper fence = 28 + 1.5 × 21 = 28 + 31.5 = 59.5. By applying  
the rule of thumb, which states that any value < lower fence OR > upper fence is an outlier. As such, 90 > 59.5,  
which confirms it as an outlier, falling in 1.45 IQR units as determined by the formula, Distance from fence =  
(Outlier value - Upper fence) / IQR translating into (90 - 59.5) / 21 = 30.5 / 21 = 1.45 IQRs beyond the fence.  
DISCUSSION  
Implication Of Results On:  
(A) Process Capability  
A median of 14 describes a process that is not likely capable and the IQR of 21 signals very high variation  
relative to the central value (Median 14). This indicates a process that is very wide spread hence unpredictable.  
Page 985  
The confirmed high variation of the CDF process means that it is unstable and therefore unpredictable. Fernades  
(2022) emphasizes that the purpose of capability analysis is to predict the performance of processes that is why  
it is critical to make sure that they are in statistical control. Otherwise, capability becomes impossible and process  
improvement becomes compromised as well, which is against the culture of operational excellence. No wonder  
Participant Acomplained that, “It is frustrating because projects take a very long time to be implemented despite  
submitting project proposals in good time.” And Respondent B intimated that sometimes implementation of  
projects takes up to 10 to 12 months. It is very clear that such delays are due to the identified high variation.  
Unfortunately, high variation also has negative consequences on resources allocation  
(B) Funding Or Resource Allocation  
High variation causes budget unreliability as fund disbursement and project completion cannot be predicted.  
Since budgets are prepared on assumptions, forecasting becomes very problematic as actual timings could vary.  
This eventually leads to under-utilization or late utilization of allocated funds, which is the status quo in the  
current CDF process in Zambia. The fact that Government allocates CDFs annually, projects meant to be  
implemented in Q1 or 2 may spill into Q3 or Q4 hence funds meant for earlier use are delayed. This is evidenced  
by the fact that there is no determined period in Zambia for projects that are funded in phases.  
RECOMMENDATION  
The local councils must prioritize and improve data collection and record keeping in order to stabilize the  
process. Usable data enables accurate forecasting and capability monitoring. This should be done by  
implementing among other things the following interventions:  
Digital tracking of every proposal application  
Timestamps for all process steps  
Dashboards for pending, approved and delayed projects  
Real-time updates on application progress  
Reduce special cause variation by monitoring long delays beyond the approved time frame  
CONCLUSION  
The quantitative analysis shows that the current CDF process is highly variable, unpredictable and statistically  
unstable. The fact that the process does not follow a normal probability distribution, proves that most projects  
do not meet the expected time requirements and fund disbursements are unpredictable. These inefficiencies  
negatively affect budgeting and resource allocation among other things, and overall project implementation  
timelines. As such, improving process stability is vital before meaningful capability analysis can be performed.  
Ethical Considerations  
The study was approved on 11th April, 2025 by humanities and social science ethics committee at the University  
of Zambia. Human participants and Respondents were issued with consent forms, which they signed before  
answering questions.  
Funding: The study was not funded  
Conflict Of Interest: The Author(s) declares no conflict of interest  
Page 986  
REFERENCES  
1. Fernandes.  
M
(2022).  
The  
importance  
of  
Capability  
Analysis.  
2. Kardas, M. and Kwilinski, A (2024). Enhancing Process Stability and Quality Management: A  
Comprehensive Analysis of Process Capability Indices. Virtual Economics, 6(4), 73-92.  
3. Montgomery, D. C (2013). Introduction to Statistical Quality Control. 7. ed. Danvers: John Wiley &  
Sons.  
4. Nagesh P, and Swamy D.R. (2014), Process Capability Analysis - A Tool for Process Performance  
measures and metrics - A Case Study: International Journal for Quality Research 8(3) 399-416 ISSN  
1800-6450  
5. Namane Zahir (2023). 7 Basics that Define Variation & Process Capability in 6 Sigma.  
Accessed 6/11/2025  
6. Peterka P (2024). Process Variation in Lean Six Sigma. Everything to Know. [Online].  
7. Six Sigma (2024). Process Variation in Lean Six Sigma. Everything to Know. [Online].  
8. Yang, C.; Yu, Q (2023). Placement and size-oriented heat dissipation optimization for antenna module in  
space solar power satellite based on interval dimension-wise method. Aerosp. Sci. Techno, 134, 108155.  
[CrossRef]  
Page 987