Assignment 1

Operations Management
Class Code: W24 BU3044-SCM G1

Prof: Denton Dewar

Team Members:
Anges Martin 202207143
Pardeep 202302938
Mohit Kumar 202303296
Gurpreet Singh 202302582
Shubham 202203359

Due Date: Feb 8, 2024

1Q. Explain the basic concepts of forecasting and time series in an operations management
context
Forecasting and time series analysis are critical operations management techniques for planning,
decision-making, and resource allocation. Following are the explanations of the fundamental topics of
forecasting and time series:

Forecasting:
Forecasting is the procedure of predicting what will happen in the future by considering events from the
past and the present. In simple terms, it is a decision-making tool that analyzes previous data and
patterns to help companies deal with the effect of future uncertainty. It is a planning tool that allows
businesses to plan out their future steps and develop budgets that, probably, will cover any potential
risks.

Qualitative forecasting:

Qualitative forecasting is a way of generating financial projections based on expert judgment. Employees
who are experts in qualitative forecasting find and analyze the link between previous operational
information and projected future operations. This enables the experts to make predictions about how a
firm will do in the future based on their opinions and data gathered from other sources, such as
employee surveys or market research.

Quantitative Methods:

Quantitative forecasting is a data-driven approach that allows businesses and salespeople to make
accurate forecasts about future decisions. It analyzes prior sales data to determine performance and
areas for improvement. By observing patterns over time, firms can make smarter decisions about
realigning strategy for consistent outcomes. Quantitative forecasting also helps organizations determine
whether to take risks and how to enhance their operations. Overall, it provides a deeper understanding
of a company's activities, allowing for more effective and meaningful decision-making.
Time Series:

Time series approaches use historical data to forecast future events. They are a collection of discrete-
time measurements made at various points in time. Sunspot counts, ocean tidewater heights, and the
Dow Jones Industrial Average are among examples. Line charts are commonly used to plot time series
data in a variety of domains, including statistics, signal processing, pattern identification, econometrics,
mathematical finance, weather forecasting, earthquake prediction, and control engineering.

Trend: A trend is the constant increasing or downward movement of demand. This could be related to a
product's life cycle.

Cycle: A cycle is a pattern in data that lasts longer than a year. They are frequently linked to events such
as interest rates, political climate, consumer confidence, and other market variables.

Seasonal: Many items follow a seasonal pattern, which is characterized by predictable changes in
demand that occur year after year. Seasonality has a significant impact on both fashion and sporting
items.

Irregular variations: Often, demand is impacted by an incident or set of occurrences that are unlikely to
occur again. Extreme weather, a strike on a college campus, or a power outage are all potential
examples.

Random variations: Random variations are inexplicable fluctuations in demand that persist after all
other factors have been evaluated. This is commonly referred to as noise.

Accuracy Evaluation: After creating forecasts, it is critical to analyze their accuracy to determine the
dependability of the predictions. Metrics used to assess accuracy include Mean Absolute Error (MAE),
Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and forecast bias.

Forecasting is a decision-making tool that uses past and present events to predict future events. It
involves qualitative and quantitative methods, using expert judgment and historical data. Time series
approaches analyze trends, cycles, seasonality, irregular variations, and random variations, with accuracy
evaluation crucial for determining predictions' dependability.

2Q. Explain how to apply simple moving averages and exponential smoothing models. Use
numerical examples.
Simple Moving Average:

The computation of the Simple Moving Average entails dividing the cumulative values over a specific
interval by the entire number of periods. For long-term trend predictions, this is quite helpful. However,
it may take longer to respond to abrupt or fast fluctuations in pricing.

How to find out simple moving average: -

The equation for the simple moving average:

Day(n) Price (P)

1 $10
2 $12
3 $11
4 $16
5 $15
6 $8
7 $12
8 $16
9 $14
10 $17
Let’s give an example, suppose a company posted its closing stock prices:

Using a 5-day SMA, the 5-day SMA is $12.8.

SMA = (10+12+11+16+15) / 5 = 12.8

Using a 10-day SMA, the 10-day SMA is $13.1.

SMA = (10+12+11+16+15+8+12+16+14+17) / 10= 13.1.

Based on these calculations, it is evident that the 10-day Simple Moving Average (SMA) is somewhat
greater than the 5-day SMA on Day 10, with values of $13.1 and $12.8 respectively. This indicates that
the short-term moving average (5 days) is lower than the long-term moving average (10 days) at this
point. This could suggest a short-term downtrend in the stock's price, as recent prices are weighted more
heavily in the 5-day SMA calculation compared to the 10-day SMA. However, it's important to note that
these moving averages alone may not provide a complete picture of the stock's trend, and other factors
should be considered in conjunction with these values for a comprehensive analysis.
Exponential Smoothing: -

Exponential smoothing is a technique used to predict future values of a single variable time series data.
Time series approaches operate based on the idea that a forecast is a calculated sum of previous
observations or time lags, with each component being assigned a certain weight. The Exponential
Smoothing time series approach operates by allocating exponentially diminishing weights to previous
observations. The term "exponentially decreased" is used to describe the gradual weight reduction
allocated to each demand observation. The model presupposes that the future will exhibit a degree of
similarity to the immediate past.

The formula for calculating exponential smoothing:

Ft+1 = α⋅Yt+(1−α) ⋅Ft

 Ft+1 is forecast for the next period.
 Yt is actual observation for the at the time.
 Ft is the forecast for the current period.
 And α is the smoothing factor, value of α is always greater than 0 and less than 1.

Example:

Assume you possess a time series dataset that depicts the monthly sales of a product during the recent
months:

Month Sales
Jan 100
Feb 120
Mar 130
Apr 140
May 150

Firstly, an initial forecast value is required. This could represent the initial recorded value, the mean of
the initial set of observations, or any other plausible estimation. Let's adopt the sales value for January
(100) as our starting projection.

the value of the smoothing constant is between 0 and 1 which defines the weight given to the most
recent observation. A larger value of α assigns greater importance to recent observations. The selection
of α is contingent upon the characteristics of the data and the intended degree of smoothing. Let's
choose α = 0.2 for this case.

Now we will use the exponential smoothing formula:

Forecast for June= α * (Actual Sales for May) + (1 - α) * (Previous Forecast)

Forecast for June= 0.2 * 150 + 0.8 * 100

=30+80

=110
Therefore, using exponential smoothing the forecast for June is 110.

The procedure can be iterated for consecutive months, employing the projected value for the present
month as the preceding projection for the following month. The values of the smoothing constant α and
the beginning forecast value can be modified according to the unique requirements and peculiarities of
the data.

3Q. Explain how to apply regression and the role of judgment in forecasting.
In regression analysis, the relationship between a dependent variable and one or more
independent variables is modeled using statistical techniques. Regression aids in forecasting by
making predictions about future values based on trends in historical data. Regression analysis
involves model fitting, variable selection, and accuracy evaluation. Nonetheless, judgment plays
a vital role. Analysts must choose which variables to include while taking contextual awareness
and subject expertise into account. In addition to helping to discover outliers and validate
model assumptions, judgment also directs the interpretation of outcomes. The knowledge of
the analyst guarantees that the regression model is in line with the particular forecasting
objectives, preventing a slavish dependence on statistical results and encouraging a
sophisticated, well-informed method of forecasting.
Moreover, judgment is essential in managing the drawbacks of regression, including possible
multicollinearity and non-linear correlations. When deciding on model complexity, analysts
must weigh simplicity against accuracy. Furthermore, predicting frequently incorporates
unpredictably occurring external events; in these situations, judgment helps add qualitative
insights and modify projections. Forecasting is further improved by being aware of the business
environment, market dynamics, and any disruptions. To put it simply, regression offers a
quantitative foundation, but human judgment guarantees a thorough, flexible, and context-
aware forecasting method that transcends statistical mechanics. Regression in forecasting
scenarios is more reliable and applicable when it combines human understanding and analytical
rigor.
Also, using judgment is essential when deciding on the right predicting time range. Analysts
have to take into account the significance of past data, possible seasonality, and the influence of
outside events. A sophisticated grasp of the issue is necessary to modify the model in response
to shifting circumstances and recognize instances in which historical trends can diverge. To
successfully communicate uncertainties and constraints, results communication also requires
judgment. Involving stakeholders and comprehending the organizational context is crucial to
making sure the prediction is in line with larger business plans. Regression is a useful tool, but
human judgment applied with consideration transforms forecasting from a statistical exercise
into a process that is both strategic and perceptive.
Additionally, judgment plays a part in model validation and ongoing improvement. By using
methods like cross-validation, analysts may evaluate the regression model's dependability and
determine when changes are necessary. To make sure the model fits fresh data adequately, they
balance the risk of underfitting and overfitting. The investigation of alternative theories and the
assessment of potential biases in the data are guided by judgment. Since predicting is by its very
nature uncertain, it is essential to include professional judgment and subjective views. To ensure
that forecasts remain relevant and accurate in constantly changing contexts, forecasting success
necessitates an iterative process in which analysts apply their judgment to modify models in
response to feedback from the real world. Regression in forecasting is essentially more
adaptive, robust, and applicable to real-world situations when it incorporates human judgment.
4Q. Explain the concept of capacity - include the principles and logic of the theory.
In the framework of operations management, capacity. This core idea is essential for
streamlining procedures, allocating resources, and enhancing overall organizational
effectiveness. A network, process, or resource's capacity is the highest output or burden it can
manage in a specific amount of time. It includes both real and intangible components, such as
service delivery, manufacturing lines, and human capacities. An organization that practices good
capacity management may meet demand while being cost-effective.
Capacity Management Fundamentals:
Prediction and Planning: Predicting future demand is a key component of capacity
planning. To estimate resource needs, organizations examine market trends, seasonality,
and historical data. Anticipating changes is where the reasoning lies. A retailer needs to
budget for extra capacity, for example, during the holiday season.
I. Expenses and Alternatives: Trade-offs are a part of decisions regarding capacity. While
purchasing more machinery boosts output, it comes at a price.
-It is the idea to assess these trade-offs. Is it better for a hospital to add more beds or make
improvements to its client flow monitoring systems?
II. Demand as well as Supply Balance: - Capacity should either equal or exceed demand,
according to logic. Unfulfilled desire is the outcome of undercapacity, and inefficiency is
the result of overcapacity.
-Capacity is adjusted by organizations according to peak and off-peak times. Airlines, for
instance, double their capacity during the summer months.
III. Material use: Here, maximizing resource use is the guiding concept. Businesses need to
find a balance between overusing (ineffectiveness) and underusing (wasted resources).
-It makes sense that allocating resources efficiently results in the best possible output. This is
relevant to call centers, manufacturing facilities, transportation systems, and other areas.
IV. Personal capability: - Human capability is an important factor that goes beyond material
resources. Skills, knowledge, and flexibility are all part of it. Through education and skill-
building initiatives, human potential is to be developed and nurtured.
 V. Network Capacity: - Software, databases, and networks are examples of technological
systems that have capacity restrictions of their own.
-To manage the amount of data, user usage, and computing needs, logic resides in evaluating
and improving system capacity.
Capacity Optimization's Logical Basis
i. Efficiency: Maximizing production while utilizing the fewest resources is the goal of
organizations. Profitability is contingent upon efficiency. Utilizing resources efficiently
should save costs and provide businesses with a competitive edge.
ii. Flexibility: Flexibility is taken into account in capacity planning. Companies have to
change to meet shifting consumer demands. Integrating flexibility into procedures to
enable prompt modifications when necessary is the guiding idea.
iii. Risk management: The logic in risk management is to reduce the hazards brought on by
excesses or shortfalls of capacity. When creating backup plans, organizations evaluate
the risks. Hospitals, for instance, are ready for unexpected spikes in patient traffic.
iv. Gains in Capacity: Economies of scale are frequently the result of greater capacity. Per-
unit expenses are lowered by bulk production. Utilizing scale benefits while preserving
quality is the guiding idea.
v. Satisfaction of Customers: It stands to reason that satisfying demands results in happier
customers. Service level agreements (SLAs) and capacity are matched by organizations
to guarantee prompt delivery.
Real-world Illustrations
a) Production: Automobile producers maximize the capacity of their assembly lines to
reach their goals.
Logic: Production lines with high efficiency save expenses while raising quality.
b) Medical Care: - To handle patient admissions, hospitals control bed capacity. The idea is
to balance capacity to assure patient care and avoid overcrowding.
c) Mobility: - Seasonal demand informs airlines' adjustments to flight capacity.
Rationale: Minimize vacant seats while satisfying passenger demands.
In conclusion, risk assessment, resource allocation, and strategic decision-making are all part of
capacity management. Achieving effectiveness, adaptability, and client happiness is the
rationale. Organizations can prosper in dynamic contexts by grasping the ideas of capacity.
Recall that capacity is dynamic and changes in response to new situations. Resilience and
flexibility are guaranteed by effective management, ensuring long-term operational success.
 References

Divyant, A. (2022 May 08) Introduction to Time Series. https://www.linkedin.com/pulse/time-series-part-

1-introduction-divyant-agarwal/

ECDPM (2008) Capacity Change and Performance: Insights and Implications for Development
Cooperation, Chapter 3: The Concept of Capacity – introduction section on page 23

Ernita M.J & Thomas J. (2021, September 03) The Concept of Capacity.
https://link.springer.com/chapter/10.1007/978-3-030-80564-7_3

Freeson, O. (2023, August 24) Mastering Business Forecasting

https://www.linkedin.com/pulse/mastering-business-forecasting-time-series-analysis-o-pmp-csm-

Rob, J. H. and George, A. (2018) Forecasting: Principles and Practice (2nd ed). Chapter 3 Judgement
Forecasting.

Rob, J. H. and George, A. (2018) Forecasting: Principles and Practice (2nd ed). Chapter 5 Time Regression.

Robert, N. (2024) Statistical forecasting: Notes on regression and time series analysis
https://people.duke.edu/~rnau/411avg.htm

Saylor Academy. (2019, November 4). Principles of Marketing. Forecasting.

https://saylordotorg.github.io/text_principles-of-marketing-v2.0/s19-03-forecasting.html

Will K. (2021, January 24) How Does a Company Maximize Output?

https://www.investopedia.com/terms/c/capacity.asp