Please enable JavaScript.
Coggle requires JavaScript to display documents.
data sience - Coggle Diagram
data sience
kep points
how tdata transfer
shop online
navigate
DPDHL 100,hunders of data
resource
imporve our business
help induivulas
incrase effeicndy
business growth
how its work
data scinence at DPDHL
Hr
excited
more 10000 application
tracking system
improve effeicicy
serach engine
sorting system
more imorovemnet
how
look on data
charterstick
build agorithum
150,000 , integrat with our system
problem
data topoin
rely of our effecincy
more transperncy
to be
emplouer of choice
what we can imporve
in
sustinable way
reduce co2
6 june 2024
key
hisotory to predice
error in the data
large amount not make sence
differnet type of useful data
noticd
if there is holiday
update the algorithum
help us to predice
need horistical
change the porcess to get the data
data need to be orgnaize
summary
start with concerte
good quality
context
knowldeg
clean
chapter 4
help
risk
6 turks
3 truks / relaity
last minute
conutine devlpting
modeling
what is the best model
forecast mdel
implented by IT
can
levarage
gain insight
idnitfy pattern
suporot diecison
preidct outcome
achinve
combine
data scinece forecasting
cant
100% alogourithm
without context
carete value
predict 100%
where
optimize
large amount of data
positive imapct
more preditciton
feel
prode
daily
limiteaion
adapt
IT enable the model
aglogiriutm
set of insuttions
to pc
to solve an issue
keep
clean
pahtone coder / all
all can contribute
inpute
open mindset
what is all about
locsus 5 year
not secedul enough
if ,
not all
cause dealy
here about his challenge
estimate
hard to estimate
if over estimage less efficency
use his intitions
18 errror date
making decsion
more data better dision make
data
how many truck
preivous week
discripbe what happen last week
how ?
calssica lapprocah
reporting
wide of datset
weather or holiday
use algorithoum
set of insoutcions for pc to sovle problem
valuble insight
refer to model
machine learn
learn from past dat
historical transportation
crearing remmandation
AI , yes
cognitive task
online product recommadn
partiacly
humman experice unique
partially
summery
ensure work effecint
happy with result
all customer and employees
using advance algorithum
more accurate
many limtiates and need our experiance
forecasting transport volumes
best service
big data
high data being produced
hugme amount of data generation
qulity more than quantity
how its work
how operaiton work
what the data been gernetatd
clean data
test and improve model
forecasting reduce to 8%
integarte with IT
data sience forecast
add value
support DHL strategy
chapter 2
key point
start with the customer
optimze the porcess
learn from past dat
plan our resources
our capcity
how many trucks / aircart
rooting of the last mile
deilvery
supoprt functions
predict event
our recuritor find bes application
custome quote
everthing in place
greate service
process
hard
why the porcess is like this
receive emials
process manual
go back to the customers
taking days
idenify
detect the infomrations
so much faster
we cas use it for pricing
asses sucssus ration
quote from to , you may say
may lower the price
genert info withing minute
supporting functions
over 50 % late
finance
not always get pay in time
disagree for our charges
customer in advance
need to pritotirez
memory of the collector
we will use machine learning
learing from the past how csutomer pay
tell the collecter forget about them
happey with model
to start
concerte question
what is the best route
where the customer pay invoice
need
business knowledge
high quality of data
data
pictue
words
numbers
500,000+
scan
customer in touch
email
where data stored
data base
data warehoue
corprate data lake