Date post: | 19-Aug-2018 |
Category: |
Documents |
Upload: | truongdieu |
View: | 218 times |
Download: | 0 times |
Understanding and Improving the
Supermarket Price Reduction Process
Duncan Apthorp
Supply Chain Development
Supply Chain Development Projects Improving promotions
Replacing our sales forecasting
Reduced
Waste
Stock
Predicting weather effects on sales
Less
summer
food waste
Better
on-shelf
availability
Optimising store operations
Less
Waste
Where does Matlab fit in Tesco?
• Runs the business
• Hard Real Time
• 24/7/365 uptime
• Monthly updates
•12 month lead time
IBM Mainframe Teradata Data Warehouse
• 5 years of data
•100 Tbytes, 150 cores
• Soft real time
• SQL Only for now
• Batch jobs / user queries
• Agile Development
• Analysis
• Model Development
• Simulations
Matlab
Desktop and Servers
Why reduce products going out of date?
• It’s good for our customers
• It’s good for the environment
• It’s good for business
• It’s a legal requirement
What is the Process ?
Up to 2010
• Products going out of date are scanned each evening
• Reduced up to 3 times
• Expiring product taken off sale before midnight
• Reduction percentages based on colleague’s experience
What is the Process ?
2010 – Automated Reduction Percentages
• Reduction percentages automated
• Reductions calculated automatically:
• Quantity going out of date
• Sales forecast
• Product type
The automated process brought major benefits
0%
100%
200%
300%
400%
500%
0% 25% 50% 75% 100%
Sale
s U
plift
Reduction
Price Elasticity
Fruit
& Veg
Meat
Increased
Sales
Less
Waste
2014 – New Reduction Model
• Second Tesco Mathworks joint development
• Tesco • Business / systems knowledge
• Big data expertise
• Mathworks • Increase in capacity
• Data Analytics – new ways of analysing and visualising data
• Statistical Modelling – new approaches
• Production model development
• MATLAB as the common language
2014 - Detailed Reduction Optimisation Model
Programme was based on learnings from previous project:
1. Define programme aims and KPIs
2. Understand the data
3. Build a measurement framework
4. Build your first models, get a quick win
5. Then build the final models
KPIs - what do we want to achieve?
• Make it simple and clear for our customers
• Minimise our impact on the environment (waste tonnes)
• Minimise the cost (waste in £)
• Minimise the effort involved for our colleagues in store
14 Apr 00:00 15 Apr 00:00 16 Apr 00:00 17 Apr 00:00 18 Apr 00:000
5
10x 10
5
£ W
aste
14 Apr 00:00 15 Apr 00:00 16 Apr 00:00 17 Apr 00:00 18 Apr 00:000
5
10x 10
5
Sto
ck
Date Scan
Overstock
Reduction
Sales
Stock, Waste
START
Evaluation Framework
Model
• Tesco retail and data knowledge
• Mathworks statistics and data analytics
• Models effect of reduction on sales rate
• Predicts KPIs
• Creates optimum reductions
Model Simplified Schematic
Merge
Scans
Sales
Stock
Test
Train Models for
Products
Select
Reduce %
Calculate
Uplift
Calculate
Quantity
Phased roll out to learn and measure benefits
Fully
Introduced
(2000+ stores)
Store director
group
(30 to 40 stores)
Nationally representative trial
(200 stores)
Customer Service Level Availability
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 98.3% 98.1% -0.2%
Control Stores: 98.4% 98.0% -0.4%
Dot Com Availability
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 96.7% 96.3% -0.4%
Control Stores: 96.7% 96.0% -0.8%
AM Gap Scan Availability Unweighted
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 98.8% 98.4% -0.4%
Control Stores: 98.8% 98.5% -0.3%
AM Gap Scan Availability Weighted
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 98.5% 98.2% -0.3%
Control Stores: 98.5% 98.2% -0.4%
PM Gap Scan Availability Unweighted
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 98.6% 98.4% -0.2%
Control Stores: 98.6% 98.4% -0.2%
PM Gap Scan Availability Weighted
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 97.3% 97.2% -0.1%
Control Stores: 97.4% 97.1% -0.3%
Stock Holding as Days Cover
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 9.7 9.9 0.2
Control Stores: 9.1 9.6 0.4
Back room Stock Holding as Days Cover
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 3.4 3.4 0.0
Control Stores: 3.2 3.4 0.2
Sales Forecast error
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 58.6% 54.0% -4.6%
Control Stores: 55.7% 56.3% 0.6%
Sales Forecast Bias (Adj sales vs 0hr Forecast)
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: -14.5% -6.2% 8.3%
Control Stores: -12.7% -13.5% -0.8%
DC to store service level
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 97.4% 95.8% -1.6%
Control Stores: 97.4% 97.1% -0.4%
Trial vs Control
Performance
-1.2%
Trial vs Control
Performance
-5.2%
Trial vs Control
Performance
9.0%
Trial vs Control
Performance
0.14%
Trial vs Control
Performance
-0.17
-0.25
Trial vs Control
Performance
BSF Grocery National Roll-Out
(All)
Trial vs Control
Performance
-0.01%
Trial vs Control
Performance
0.25%
Trial vs Control
Performance
Trial vs Control
Performance
0.09%
0.31%
0.04%
Trial vs Control
Performance
96.5%
97.0%
97.5%
98.0%
98.5%
99.0%
99.5%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
90%
91%
92%
93%
94%
95%
96%
97%
98%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
97.6%
97.8%
98.0%
98.2%
98.4%
98.6%
98.8%
99.0%
16 A
pr
20 A
pr
24 A
pr
28 A
pr
02 M
ay
06 M
ay
10 M
ay
14 M
ay
18 M
ay
22 M
ay
26 M
ay
13 J
un
17 J
un
21 J
un
25 J
un
29 J
un
03 J
ul
07 J
ul
11 J
ul
15 J
ul
19 J
ul
23 J
ul
27 J
ul
31 J
ul
04 A
ug
08 A
ug
12 A
ug
96.0%
96.5%
97.0%
97.5%
98.0%
98.5%
99.0%
99.5%
100.0%
100.5%
16 A
pr
20 A
pr
24 A
pr
28 A
pr
02 M
ay
06 M
ay
10 M
ay
14 M
ay
18 M
ay
22 M
ay
26 M
ay
13 J
un
17 J
un
21 J
un
25 J
un
29 J
un
03 J
ul
07 J
ul
11 J
ul
15 J
ul
19 J
ul
23 J
ul
27 J
ul
31 J
ul
04 A
ug
08 A
ug
12 A
ug
95.5%96.0%96.5%97.0%97.5%98.0%98.5%99.0%99.5%
100.0%100.5%
16 A
pr
20 A
pr
24 A
pr
28 A
pr
02 M
ay
06 M
ay
10 M
ay
14 M
ay
18 M
ay
22 M
ay
26 M
ay
13 J
un
17 J
un
21 J
un
25 J
un
29 J
un
03 J
ul
07 J
ul
11 J
ul
15 J
ul
19 J
ul
23 J
ul
27 J
ul
31 J
ul
04 A
ug
08 A
ug
12 A
ug
95.0%
95.5%
96.0%
96.5%
97.0%
97.5%
98.0%
98.5%
16 A
pr
20 A
pr
24 A
pr
28 A
pr
02 M
ay
06 M
ay
10 M
ay
14 M
ay
18 M
ay
22 M
ay
26 M
ay
13 J
un
17 J
un
21 J
un
25 J
un
29 J
un
03 J
ul
07 J
ul
11 J
ul
15 J
ul
19 J
ul
23 J
ul
27 J
ul
31 J
ul
04 A
ug
08 A
ug
12 A
ug
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
-35%
-30%
-25%
-20%
-15%
-10%
-5%
0%
5%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
70%
75%
80%
85%
90%
95%
100%
105%
110%
115%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
Customer Service Level Availability
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 98.3% 98.1% -0.2%
Control Stores: 98.4% 98.0% -0.4%
Dot Com Availability
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 96.7% 96.3% -0.4%
Control Stores: 96.7% 96.0% -0.8%
AM Gap Scan Availability Unweighted
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 98.8% 98.4% -0.4%
Control Stores: 98.8% 98.5% -0.3%
AM Gap Scan Availability Weighted
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 98.5% 98.2% -0.3%
Control Stores: 98.5% 98.2% -0.4%
PM Gap Scan Availability Unweighted
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 98.6% 98.4% -0.2%
Control Stores: 98.6% 98.4% -0.2%
PM Gap Scan Availability Weighted
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 97.3% 97.2% -0.1%
Control Stores: 97.4% 97.1% -0.3%
Stock Holding as Days Cover
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 9.7 9.9 0.2
Control Stores: 9.1 9.6 0.4
Back room Stock Holding as Days Cover
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 3.4 3.4 0.0
Control Stores: 3.2 3.4 0.2
Sales Forecast error
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 58.6% 54.0% -4.6%
Control Stores: 55.7% 56.3% 0.6%
Sales Forecast Bias (Adj sales vs 0hr Forecast)
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: -14.5% -6.2% 8.3%
Control Stores: -12.7% -13.5% -0.8%
DC to store service level
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 97.4% 95.8% -1.6%
Control Stores: 97.4% 97.1% -0.4%
Trial vs Control
Performance
-1.2%
Trial vs Control
Performance
-5.2%
Trial vs Control
Performance
9.0%
Trial vs Control
Performance
0.14%
Trial vs Control
Performance
-0.17
-0.25
Trial vs Control
Performance
BSF Grocery National Roll-Out
(All)
Trial vs Control
Performance
-0.01%
Trial vs Control
Performance
0.25%
Trial vs Control
Performance
Trial vs Control
Performance
0.09%
0.31%
0.04%
Trial vs Control
Performance
96.5%
97.0%
97.5%
98.0%
98.5%
99.0%
99.5%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
90%
91%
92%
93%
94%
95%
96%
97%
98%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
97.6%
97.8%
98.0%
98.2%
98.4%
98.6%
98.8%
99.0%
16 A
pr
20 A
pr
24 A
pr
28 A
pr
02 M
ay
06 M
ay
10 M
ay
14 M
ay
18 M
ay
22 M
ay
26 M
ay
13 J
un
17 J
un
21 J
un
25 J
un
29 J
un
03 J
ul
07 J
ul
11 J
ul
15 J
ul
19 J
ul
23 J
ul
27 J
ul
31 J
ul
04 A
ug
08 A
ug
12 A
ug
96.0%
96.5%
97.0%
97.5%
98.0%
98.5%
99.0%
99.5%
100.0%
100.5%
16 A
pr
20 A
pr
24 A
pr
28 A
pr
02 M
ay
06 M
ay
10 M
ay
14 M
ay
18 M
ay
22 M
ay
26 M
ay
13 J
un
17 J
un
21 J
un
25 J
un
29 J
un
03 J
ul
07 J
ul
11 J
ul
15 J
ul
19 J
ul
23 J
ul
27 J
ul
31 J
ul
04 A
ug
08 A
ug
12 A
ug
95.5%96.0%96.5%97.0%97.5%98.0%98.5%99.0%99.5%
100.0%100.5%
16 A
pr
20 A
pr
24 A
pr
28 A
pr
02 M
ay
06 M
ay
10 M
ay
14 M
ay
18 M
ay
22 M
ay
26 M
ay
13 J
un
17 J
un
21 J
un
25 J
un
29 J
un
03 J
ul
07 J
ul
11 J
ul
15 J
ul
19 J
ul
23 J
ul
27 J
ul
31 J
ul
04 A
ug
08 A
ug
12 A
ug
95.0%
95.5%
96.0%
96.5%
97.0%
97.5%
98.0%
98.5%
16 A
pr
20 A
pr
24 A
pr
28 A
pr
02 M
ay
06 M
ay
10 M
ay
14 M
ay
18 M
ay
22 M
ay
26 M
ay
13 J
un
17 J
un
21 J
un
25 J
un
29 J
un
03 J
ul
07 J
ul
11 J
ul
15 J
ul
19 J
ul
23 J
ul
27 J
ul
31 J
ul
04 A
ug
08 A
ug
12 A
ug
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
-35%
-30%
-25%
-20%
-15%
-10%
-5%
0%
5%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
70%
75%
80%
85%
90%
95%
100%
105%
110%
115%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
Customer Service Level Availability
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 98.3% 98.1% -0.2%
Control Stores: 98.4% 98.0% -0.4%
Dot Com Availability
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 96.7% 96.3% -0.4%
Control Stores: 96.7% 96.0% -0.8%
AM Gap Scan Availability Unweighted
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 98.8% 98.4% -0.4%
Control Stores: 98.8% 98.5% -0.3%
AM Gap Scan Availability Weighted
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 98.5% 98.2% -0.3%
Control Stores: 98.5% 98.2% -0.4%
PM Gap Scan Availability Unweighted
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 98.6% 98.4% -0.2%
Control Stores: 98.6% 98.4% -0.2%
PM Gap Scan Availability Weighted
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 97.3% 97.2% -0.1%
Control Stores: 97.4% 97.1% -0.3%
Stock Holding as Days Cover
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 9.7 9.9 0.2
Control Stores: 9.1 9.6 0.4
Back room Stock Holding as Days Cover
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 3.4 3.4 0.0
Control Stores: 3.2 3.4 0.2
Sales Forecast error
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 58.6% 54.0% -4.6%
Control Stores: 55.7% 56.3% 0.6%
Sales Forecast Bias (Adj sales vs 0hr Forecast)
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: -14.5% -6.2% 8.3%
Control Stores: -12.7% -13.5% -0.8%
DC to store service level
Pre Trial
Cumulative
Trial
Cumulative
Trial vs Pre
Trial
Trial Stores: 97.4% 95.8% -1.6%
Control Stores: 97.4% 97.1% -0.4%
Trial vs Control
Performance
-1.2%
Trial vs Control
Performance
-5.2%
Trial vs Control
Performance
9.0%
Trial vs Control
Performance
0.14%
Trial vs Control
Performance
-0.17
-0.25
Trial vs Control
Performance
BSF Grocery National Roll-Out
(All)
Trial vs Control
Performance
-0.01%
Trial vs Control
Performance
0.25%
Trial vs Control
Performance
Trial vs Control
Performance
0.09%
0.31%
0.04%
Trial vs Control
Performance
96.5%
97.0%
97.5%
98.0%
98.5%
99.0%
99.5%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
90%
91%
92%
93%
94%
95%
96%
97%
98%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
97.6%
97.8%
98.0%
98.2%
98.4%
98.6%
98.8%
99.0%
16 A
pr
20 A
pr
24 A
pr
28 A
pr
02 M
ay
06 M
ay
10 M
ay
14 M
ay
18 M
ay
22 M
ay
26 M
ay
13 J
un
17 J
un
21 J
un
25 J
un
29 J
un
03 J
ul
07 J
ul
11 J
ul
15 J
ul
19 J
ul
23 J
ul
27 J
ul
31 J
ul
04 A
ug
08 A
ug
12 A
ug
96.0%
96.5%
97.0%
97.5%
98.0%
98.5%
99.0%
99.5%
100.0%
100.5%
16 A
pr
20 A
pr
24 A
pr
28 A
pr
02 M
ay
06 M
ay
10 M
ay
14 M
ay
18 M
ay
22 M
ay
26 M
ay
13 J
un
17 J
un
21 J
un
25 J
un
29 J
un
03 J
ul
07 J
ul
11 J
ul
15 J
ul
19 J
ul
23 J
ul
27 J
ul
31 J
ul
04 A
ug
08 A
ug
12 A
ug
95.5%96.0%96.5%97.0%97.5%98.0%98.5%99.0%99.5%
100.0%100.5%
16 A
pr
20 A
pr
24 A
pr
28 A
pr
02 M
ay
06 M
ay
10 M
ay
14 M
ay
18 M
ay
22 M
ay
26 M
ay
13 J
un
17 J
un
21 J
un
25 J
un
29 J
un
03 J
ul
07 J
ul
11 J
ul
15 J
ul
19 J
ul
23 J
ul
27 J
ul
31 J
ul
04 A
ug
08 A
ug
12 A
ug
95.0%
95.5%
96.0%
96.5%
97.0%
97.5%
98.0%
98.5%
16 A
pr
20 A
pr
24 A
pr
28 A
pr
02 M
ay
06 M
ay
10 M
ay
14 M
ay
18 M
ay
22 M
ay
26 M
ay
13 J
un
17 J
un
21 J
un
25 J
un
29 J
un
03 J
ul
07 J
ul
11 J
ul
15 J
ul
19 J
ul
23 J
ul
27 J
ul
31 J
ul
04 A
ug
08 A
ug
12 A
ug
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
-35%
-30%
-25%
-20%
-15%
-10%
-5%
0%
5%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
70%
75%
80%
85%
90%
95%
100%
105%
110%
115%
16 A
pr
19 A
pr
22 A
pr
25 A
pr
28 A
pr
01 M
ay
04 M
ay
07 M
ay
10 M
ay
13 M
ay
16 M
ay
19 M
ay
22 M
ay
25 M
ay
11 J
un
14 J
un
17 J
un
20 J
un
23 J
un
26 J
un
29 J
un
02 J
ul
05 J
ul
08 J
ul
11 J
ul
14 J
ul
17 J
ul
20 J
ul
23 J
ul
26 J
ul
29 J
ul
01 A
ug
04 A
ug
07 A
ug
10 A
ug
13 A
ug
Automated daily project tracking
Availability
Stock Holding
Back Room Stock
Forecast Accuracy
Coming to
your local
store 2015
Working with Mathworks – some tips
• Agree the goals, and how to measure them
• Make it a joint development
• Have a single contact for day to day operations
• Hold regular high level reviews
• Don’t accept things that feel wrong
The Future – in database analytics
IBM Mainframe Teradata Data Warehouse
• in database analytics for
heavy lifting
Matlab for:
• Control
• Simulation
• Small models
Matlab
Desktop and Servers