You need to index with str which is apply for each value of Series:
data['Order_Date'] = data['Shipment ID'].str[:8]
For better performance if no NaNs values:
data['Order_Date'] = [x[:8] for x in data['Shipment ID']]
print (data)
Shipment ID Order_Date
0 20180504-S-20000 20180504
1 20180514-S-20537 20180514
2 20180514-S-20541 20180514
3 20180514-S-20644 20180514
4 20180514-S-20644 20180514
5 20180516-S-20009 20180516
6 20180516-S-20009 20180516
7 20180516-S-20009 20180516
8 20180516-S-20009 20180516
You can filter column by position, first N values like:
print (data['Shipment ID'][:2])
0 20180504-S-20000
1 20180514-S-20537
Name: Shipment ID, dtype: object
If you want to know more about the Data Science then do check out the following Data Science which will help you in understanding Data Science from scratch