INTRODUCTION

In this comprehensive study, the technical manager of the SME beverage producer has
been tasked with optimizing the process as well as the procedure for the facility of lemonade
production. Hence the study extends beyond the internal considerations for encompassing
external factors, notably under market demand and logistic efficiency since producers seek
to establish a central warehouse to streamline the distribution to the local supermarkets.
Assets as tangible and intangible have been strategically leveraged in enhancing the
accuracy of the forecasting and logistic operations where technical managers employ the
quantitative forecasting method, the “Center of Gravity approach”. The study
contemplates ways of investing in logistics optimization as well as process streamlining.
Ultimately exact output referred to a variety of lemonade flavors has been central to this
study.

2. TASK A A-DEMAND FORECASTING

Forecasting as integral to the planning process mostly aids in decision-making through the
prediction of future events. Despite the uncertainty, structured approaches may improve
success with consideration of the most available resources and the study objectives.

2.1 Forecasting methods

Organizations mostly employ qualitative and quantitative methods of forecasting for future
estimates. Qualitatively, Delphi, Expert Opinion, and Technique of the market survey are
used while quantitatively different methods such as “Simple moving average”,
“Exponential smoothing” and the “Holt Method” are used in strategic planning (Seyedan
and Mafakheri, 2020). Combining both approaches may yield the most satisfactory
predictions as ideal to the beverage producers. As a technical manager, lemonade demand-
based data getting leveraged facilitates the application of different quantitative methods in
planning for the whole upcoming year very effectively (Abbasimehr et al. 2020).

Figure 1: Comparison of the Qualitative and Quantitative methods

(Source: Abbasimehr et al. 2020)

2.2. “Prognosis values with diverse types of lemonade”

Demand for 1st two quarters of 2019 has been seen to be forecasted by the use of given
data regarding actual demand for 3 quarters of 2018.

1
2.2.1. “Lemon - Simple Moving Average”

“Simple moving average” as a statistical method of analysis has been used with time-series
data for smoothing out the fluctuations in the short-term and the long-term trends as well
(Swari et al. 2021).
“Moving Average” = Σ “Demand in the previous n periods” / n
Using the demand of first 3 quarters of 2018, the fourth quarter is calculated as,
Table 2: “Demand Forecast of the beverage type Lemon”
Calculation of the prognosis value of the type of “Lemon’ using Moving Average, considered
n=3

2.2.2. “Elderberry - First-Order Exponential Smoothing”

“First-order Exponential Smoothing” refers to Time-series forecasting for the single-variable

data excluding trends and the seasons and lacks cyclical prediction (Pan et al. 2019).
The formula followed is,
“Forecast for the current period”
= “Forecast for a previous period” + α (Demand of the Previous period – Forecast of
Previous period)
Where α is a smoothing constant that may take a value from 0 to 1.
Table 2: Demand Forecast of the type of beverage “Elderberry”
Calculations made upon the Prognosis value of the “Elderberry” by exponential smoothing
used can be,
“Exponential Smoothing Model” has been applied successfully to the 2nd quarter of 2019
and has been widely embraced over “Time Series Analysis” duly characterized by its
computation efficiency, simplicity, and adaptability for processing changes (Pan et al. 2019).
Renowned for providing accurate forecasts throughout different applications, it seeks
minimal level data capacity as well as computing resources that made it a practical choice.

2.2.3. Lychee – “Holt Method”

Holt's method mostly refers to having close accurate forecasts as in it considers levels and
trends. Hence the results get smoothed where users may select the best options for this type
of data to be analyzed (Lima et al. 2019).
The model explores three equations an “estimate of the current level”, an “estimate of the
trend”, and the “forecast of the period of future”
Table 3: “Demand forecast for beverage type Lychee”
Quarter I of 2019 -
Quarter II of 2019 -

2
3
TASK B – “OPTIMAL LOCATION SELECTION”

Site selection has been vital to organizations, impacting largely the operations, and customer
interaction. Hence criteria have been to include employee flexibility, economic viability, and
an engaging environment (Liu et al. 2019). The “Center of gravity” method has aided
beverage producers in making choices of optimal central warehouse locations depending
upon the linear distance-cost relationships as well as given demand data.

3.1. Center of Gravity method

The optimal location of supply of supermarkets has been chosen in keeping transport efforts
at the minimum level.

Table 4: Calculation of the optimal location coordinates through the use of the
“Center of Gravity method”
Consideration of the equations,

3.2. “Location factors not considered in the Center of Gravity Method”

While “Center of gravity” method has focused upon minimization of the transportation costs,
it totally overlooks the most vital factors impact to the distribution of th site selection. These
factors mostly incorporate competitive advantage, political influences and government,
geographical peculiarities, labor characteristics.

4. TASK C – “PRIORITY RULES FOR SCHEDULING PRODUCTION”

4.1. “First-come-first-served” (FCFS)

This is the job sequencing rule depending upon arrival time at the production site and mostly
get used in inventory valuation (FIFO) and possesses numerous drawbacks with orders
mostly posing shorter time of processing and extended waiting times (John and Millum,
2020). Customer orders get processed into orders received as well and the lateness has
been calculated through subtracting the due date from the cumulative process timing.
Table 5: Processing of the Sequence depending upon “First-come-first Served”
As calculated, the average lateness is 4.375 days while processing orders are considered as
per the FCFS approach.

4
4.2. “Shortest processing time” (SPT)

This mostly prioritizes the tasks depending upon processing time and selection of shortest
duration in the first short with proceeding over an ascending order. Hence processing time is
weighted over accommodating changes come to job importance. Hence customer orders
have been ranked through processing time, orders made with shortest duration as
processed first (Mohmmad et al. 2020).
Table 6: Processing sequence depending upon “Shortest Processing Time (SPT)”
Through utilizing the SPT approach, the producer can be found achieving average lateness
in 0.625 days.

4.3. “Earliest due date” (EDD)

This prioritizes projects depending upon earliest due dates mostly make it suitable to the low
and medium-volume manufacturing units with variable jobs or the batch processes. The
orders are seen to be sequenced by due dates as well as job, components with earliest due
date as processed first (Gao et al. 2020).
Table 7: Processing Sequence depending on “Earliest due date (EDD)”
Using this EDD approach, the average lateness has been 6.875 days.

4.4. “Comparison of the possible sequences determined”

From applying possible sequences over the order processing taking the different priority
rules under consideration, there comes a calculation of different points necessary to have an
evaluation of the process.

Figure 2: Comparison of Sequences

Cumulative processing timing and lateness would help out comparing sequences depending
on the average time of production, maximum delay, and number of delayed orders (Tax et
al. 2020).

4.4.1. “Average Production Time”

This is the time taken in manufacturing or processing one unit of the facility in the
manufacturing facility. This has been seen to be the lowest for “Shortest processing time”
at 15.25 days due to the fact that orders are not ranked on the basis of processing time,
making ultimately the process easy and fast (He et al. 2020). For “Earliest processing
time” it is high as 21.5 days and for “First-come-first-served approach”, it is 19 days.

5
4.4.2. Maximum Delay

Maximum delays refer to excess taken time into the procedure for running as well as ending
with a resulting number of procedures as well as cycles. For EDD, the highest average latest
has been 6.875 days while the lowest in SPT has been 0.625 days (Mátyás et al. 2020).

4.4.3. Number of Delayed Orders

EDD approach refers to having 6 delayed orders due to having a focus on due dates, rather
than the processing time, dates onto which most orders are received, and quantity of orders,
making average lateness has been high in the approach (Huang et al. 2019). Since
processing time with orders has been given priority, there come only 4 delayed orders, 5
delays to the consideration of serving the first order.

5. CONCLUSION

The focus of the study has been evaluated to shed light upon demand forecasting for three
flavors, exploring both qualitative as well as quantitative methods. Quantitative techniques
are considered including simple average, Holt method, and exponential smoothing applied in
predicting demand for the first two quarters of 2019 based on historical data. It extends to
the optimization of logistics employing the “Center of Gravity approach” in the
determination of the optimal location of the central warehouse. The scheduling aspect has
been considered with different priority rules taken under consideration, significance of
grouping the production operations into different batches and prioritizing them has been
emphasized. SPT rule has been identified as the optimal for sequence planning in
November. This understanding empowers producers in optimizing processes as well as
gains a good competitive edge, emphasizing the critical role played over proper demand
forecasting in utilization of resources and the long-term planning process.