Professional Documents
Culture Documents
Titolo PRODUCTION
IN NON REPETITIVE
sottotitoloCOMPANIES
This material and what the
Milano, XX mese 20XX Professors say in class are
Alberto PORTIOLI STAUDACHER intended for didactical use
Dipartimento Ing. Gestionale only and cannot be used
Politecnico di Milano ouside such context, nor to
Dep. Management, Economics and Industrial Engineering imply professors’ specific
alberto.portioli@polimi.it believes or opinion
Research problem and objectives
• Satisfying customer orders involves both shop floor activities and office
activities.
• High variability and unpredictability in these contexts.
• Long queues between the different stages to absorb variability.
• Local efficiency logic due to the functional organization.
• High coordination efforts required to match due dates, and low integration
between the different stages.
The non-repetitive companies production planning and control
framework
POOL OF
ORDERS
SUPPLY CHAIN
Arrival of Order release Job departure /
confirmed order Job delivery
External
Design Purchasing Assembly Testing
supplier
Single stage Single stage Single stage Single stage Single stage
planning activity planning activity planning activity planning activity planning activity
• Flow
• Takt
• Pull
• Zero Defects
Future state map of non-repetitive companies
CUSTOMERS
Product
due date
MARKETING
& SALES
Release
Guidelines for achieving Lean in non-repetitive companies
• Protect the system from external variability with a buffer upstream the first stage
(in this case Design).
• Collect orders’ processing times (if the precise information is not available, use
classes of times (e.g. low/medium-low/medium/medium-high/high).
• Plan the flow in a single point (1 single actor plans the system).
• Regulate the release of orders depending on the rhythm of the activities the
along the supply chain in order to maintain the workload level in the supply
chain stable (pull production).
• Try to maintain the flow between the stages of the supply chain. In order to
maintain the sequence of orders, work to remove the causes of blocks and the
causes of variation in the sequencing of the orders (e.g. set-ups).
Guidelines for achieving Lean in non-repetitive companies
CUSTOMERS
Product
due date Buffer variability at the
beginning of the flow
MARKETING
& SALES
Release
Guidelines for achieving Lean in non-repetitive companies
CUSTOMERS
Release
Guidelines for achieving Lean in non-repetitive companies
CUSTOMERS
Product
due date
Work to maintain the flow among the
stages. Work to eliminate the causes
MARKETING
& SALES of block and variability.
Release
Traditional managerial approach
Order entry:
Accept orders in order to
maintain the System Workload
constant
Order release:
Upper bound release
(Worklaod Limiting approach)
with high Load Limits
Dispatching logic:
As to increase the local
efficiency of the single stage
Lean managerial framework
CUSTOMERS
Order entry:
Request for tender Accept orders as to maintain
and acceptance
the System Workload
constant
Tender OPERATIONS &
SUPPLY CHAIN
Release
Framework adopted (from workload control theory)
Pool
ORDER REVIEW AND of Orders’
RELEASE Supply Chain
PRODUCTION CONTROL
throughput time throughput time Delivery
CUSTOMERS
Product
due date
MARKETING
& SALES
Release
Guidelines for achieving Lean in non-repetitive companies
Latest
Release date
Three days
from today Workload to be released in the next period on:
External
Design Purchasing Supplier Assembly Testing
Workload to be
released
Day after Workload to be released at each stage:
Tomorrow (Target Supply Chain Workload – Total Initial Load)
Number of Stages
Balanced Release
(i.e. release of the same amount of workload for
each stage)
Tomorrow
Initial Load
Testing
EXTERNAL
DESIGN PURCHASING ASSEMBLY TESTING
SUPPLIER
Today Assembly
Purchasing
Design
Impact on performances
43
42
41
40 UBR- System Workload
12500 minutes
39
38 BLR - System Workload
12500 minutes
37
36
0 10 20 30 40
For Lean approch it is therefore possible to lower the level of orders in the pre-shop pool
and still have an output that is not lower than the benchmark, thus decreasing even
further the Gross Throughput Time
Workload distribution
Traditional system
HIGH OVERALL
HIGH WORKLOAD WORKLOAD LEVEL
LEVEL and LEAD TIME
LOW
WORKLOAD
LEVEL
Lean system