library(readr)
inventory <- read_csv("data/inventory.csv") Tutorial 2: From Data to a Managerial Briefing with R and Quarto
TipLearning Goals
By the end of this tutorial, you should be able to:
- Explain what Quarto is and why analysts use it
- Run R code inside a
.qmdfile and render a report - Load a small dataset and compute a few inventory KPIs
- Produce a simple table and a basic plot
- Write a manager-friendly recommendation that separates facts vs interpretation vs decision
The Business Challenge
David Jones sells a wide range of seasonal products, including fashion items that see strong swings in demand around holidays like Christmas.
Inventory decisions are especially challenging for seasonal products because:
- Demand can spike unpredictably
- Promotions and discounts influence customer behaviour
- Regional preferences can differ
- Suppliers need advance notice to adjust deliveries
Right now, David Jones is using a fixed restocking rule:
Every week, 100 units are delivered to each region, no matter what.
But with the holiday season approaching, this rigid approach might not be enough.
Management wants a one-page briefing outlining whether they should adapt their restocking rule.
Your task is to review last year’s data and advise David Jones on how to improve inventory decisions.
Prepare these Exercises before Class
Prepare these exercises before coming to class. Plan to spend 45 minutes on these exercises.
Exercise 1: Approaches to Inventory Planning
Read the article, “A New Approach to Production and Inventory Planning,” and answer the following questions in 2–3 sentences:
(a) Define the term “inventory management”. What makes inventory management complicated for seasonal products?
(b) What is one key challenge businesses face in managing inventory today?
(c) How could this apply to a retailer like David Jones, especially during busy seasons like Christmas?
Exercise 2: A First Peek at Some Data
David Jones now wants to use last year’s data to improve inventory decisions for the upcoming holiday season.
The data you have to work from are provided below:
| Week Commencing | Region | Starting Inventory | Units Delivered | Units Sold | Ending Inventory | Price (AUD) | Promotion (Y/N) |
|---|---|---|---|---|---|---|---|
| 25 Nov | Sydney | 500 | 100 | 100 | 500 | 50 | N |
| 25 Nov | Melbourne | 500 | 100 | 150 | 450 | 50 | N |
| 2 Dec | Sydney | 500 | 100 | 180 | 420 | 45 | Y |
| 2 Dec | Melbourne | 450 | 100 | 220 | 330 | 45 | Y |
| 9 Dec | Sydney | 420 | 100 | 150 | 370 | 45 | Y |
| 9 Dec | Melbourne | 330 | 100 | 100 | 330 | 45 | Y |
| 16 Dec | Sydney | 370 | 100 | 400 | 70 | 40 | Y (Christmas) |
| 16 Dec | Melbourne | 330 | 100 | 400 | 30 | 40 | Y (Christmas) |
| 23 Dec | Sydney | 70 | 100 | 120 | 50 | 50 | N |
| 23 Dec | Melbourne | 30 | 100 | 150 | -20 (backorder) | 50 | N |
Before analysts run any code, they ask a simple question:
Can we trust this dataset enough to make a decision?
Look at the David Jones data above and answer the questions below.
(a) Identify 5 potential issues in the table. For each issue, classify it as:
- 1. Will break analysis/code: formatting and/or type problems
- 2. Unclear meaning: we need a formal definition from the business
- 3. Decision risk: could bias conclusions or lead to a bad recommendation
Fill in this template:
| Issue you noticed | Category (1/2/3) | Why it matters (1 sentence) |
|---|---|---|
(b) Write one question you would email to the data owner today that would most improve your confidence in using the data to make a recommendation.
Exercise 3: Getting Set Up in Posit Cloud
Before we proceed, we need everyone working in the same environment. Complete the following steps:
Step 1: Log in to Posit Cloud
- Go to the Posit Cloud workspace link provided on the LMS.
- Log in using your University account.
Step 2: Make your own copy of the workshop project
- Locate the workshop project in Posit Cloud.
- Click Make a Copy to create your own version.
Step 3: Open the Quarto file
- In the Files pane, locate the workshop Quarto document:
tutorial_02_inventory.qmd. - Click it to open it in the editor.
Step 4: Render the document
- Find the Render button (top of the editor).
- Click Render.
If everything is set up correctly, a new tab should appear in your browser with the rendered content.
Exercise 4: R + Quarto Vocabulary Check
Answer each in one short sentence.
(a) What does <- do in R?
(b) What is a data frame. What does a row represent in our David Jones data?
(c) What is a function in R? Explain using read_csv() as an example.
(d) What is a code chunk in a Quarto .qmd file?
(e) What does Render do in Quarto?
(f) What is a package in R? Explain using readr as an example, including how to load the package into your R session.
In-Class Exercises
You will discuss these exercises in class with your peers in small groups and with your tutor. These exercises build from the exercises you have prepared above, you will get the most value from the class if you have completed those above before coming to class.
Exercise 5: Quick Recap of Preparation Work
With your group answer the following questions:
(a) In one sentence, what is the core trade-off in inventory management for a retailer like David Jones?
(b) What are the two biggest issues with the dataset presented in Exercise 2? Justify your choices.
Exercise 6: Loading and Inspecting Some Data
(a) Open the Posit Cloud Workspace for this workshop.
(b) Load the dataset by running this line of code:
(c) We want to explain the code in (b) is doing, line by line.
- What does
```{r}mean in Quarto? - What does
#| eval: falsedo? - Do you need to change the
#| eval: falsesetting to run the code when you render the file? - What does
library(readr)do? - Explain what the assignment operator
<-?, does. - Is
read_csv()a function or a variable? - What does
"data/inventory.csv"refer to?
(d) Run each of the following commands, and explain the output to your group:
class(inventory)names(inventory)head(inventory)nrow(inventory)ncol(inventory)
Exercise 7: First Steps Working with Data
Let’s compute two manager-relevant variables:
- Revenue by region, and
- Inventory risk (how low inventory gets).
TipWhat R commands will I need?
We will use the following R commands in this exercise:
- column access with
$ - logical filtering with
== - multiplication with (
*) - Summing rows of data with
sum() - Finding matches with
any()
(a) Create a new column called revenue that equals units_sold times unit_price by running the code below. Explain what the code is doing.
# Create revenue using simple arithmetic
inventory$revenue <- inventory$units_sold * inventory$unit_price(b) Run the code below to compute total revenue for both Sydney and Melbourne. Explain what step 1, step 2 and step 3 are doing.
CautionWhat does
sum() do?
Use the help operator ? followed by the name of the function you want to use to learn more about it:
?sumOn the lower right hand side of your RStudio work space the documentation that explains the function’s use will appear. Read it to learn more!
# STEP 1: EXPLAIN WHAT IS HAPPENING
syd_data <- inventory[inventory$region == "Sydney", ]
mel_data <- inventory[inventory$region == "Melbourne", ]
# STEP 2: EXPLAIN WHAT IS HAPPENING
rev_syd <- sum(syd_data$revenue)
rev_melb <- sum(mel_data$revenue)
# STEP 3: EXPLAIN WHAT IS HAPPENING
rev_syd
rev_melb(c) A backorder occurs when ending_inventory is negative.
- Does any row have
ending_inventory < 0? - If yes, display those rows.
Run the code below and explain the steps.
CautionWhat does
any() do?
Use the help operator ? followed by the name of the function you want to use to learn more about it:
?anyOn the lower right hand side of your RStudio work space the documentation that explains the function’s use will appear. Read it to learn more!
# STEP 1: EXPLAIN WHAT IS HAPPENING
any_backorders <- any(inventory$ending_inventory < 0)
# STEP 2: EXPLAIN WHAT IS HAPPENING
any_backorders
# STEP 3: EXPLAIN WHAT IS HAPPENING
backorder_rows <- inventory[inventory$ending_inventory < 0, ]
# STEP 4: EXPLAIN WHAT IS HAPPENING
backorder_rowsExercise 8: Turning Numbers into a Story (Before We Recommend Anything)
You’ve just computed a few KPIs in R (revenue, units sold, minimum ending inventory, and whether backorders occurred).
Before we jump to “what should David Jones order?”, we need to answer a more basic question:
What do the numbers actually tell us — and what do they not tell us?
(a) What are the two most important facts you learned from your KPIs?
Write them as two short bullet points (each should reference a KPI you computed).
(b) Based on those facts, what is your group’s best guess about what will happen next week if David Jones keeps doing the same thing?
(c) What is one caveat you would tell management about your answer to (b)?
Exercise 9: Planning for This Year’s Holiday Season
David Jones’ supplier is asking you to confirm delivery quantities now, so they can prepare for the season ahead.
(a) How many units should you order for Sydney and Melbourne for the first week of the holiday season?
(b) Justify your decision based on last year’s patterns and your reasoning. Consider factors like demand trends, regional differences, price effects, and promotion timing.
(c) Suppose the supplier has limited capacity and can only supply 200 units total for the first week. How would you allocate this stock between Sydney and Melbourne? What are the risks of your choice?
Exercise 10: Render Your Document
Now render your Quarto document so that your work appears in a single report.
(a) Click Render in your .qmd file.
(b) Check that your rendered document includes:
- your answers to the interpretation questions,
- the code and output from your R exercises,
- and your recommendation for David Jones.
(c) If your document does not render, identify the error message and work with your group to fix it.
(d) What is one advantage of having your analysis and your written interpretation in the same document?