02 Production Optimization

Question 1

Measurements to implement a more flexible and more efficient production

Answer 1

• Investment in modern automation technology
• Value stream optimization in assembly
• Use of collaborative lightweight robots
• Increase of production volume
• Reduction of lead time
• Targeted use of human labour

Question 2

Task of Production Improvment

Answer 2

increase production efficiency in the four target dimensions

• Variability
• Quality
• Profitability
• Speed

Question 3

Logical target square of production

Answer 3

Logical target square of production describes the mutual relationship of the four target dimensions of production improvement.

(Abbildung)

Question 4

Logical target square of production - Contradictory conflicting goals

Answer 4

Improvement in the degree of achievement of one goal leads to a deterioration in the other goal

(Examples in Summary)

Question 5

Logical target square of production - Contrary conflicting goals

Answer 5

Fulfilment of the two goals cannot be improved at the same time -> Fulfilment of one goal without deteriorating the other is possible

(Examples in Summary)

Question 6

Logical target square of production - Target Subordination

Answer 6

On of the goals might be more important in the project context of the project -> Preferred to improve this goal

(Examples in Summary)

Question 7

Logical target square of production - Target Compatibility

Answer 7

Exists when the degree of fulfilment of two targets cannot be worsened at the same time

(Examples in Summary)

Question 8

Value Stream Design

Answer 8

Method to improve Production Processes while production is running

Question 9

Elements of a Value Stream

Answer 9

o Dotted lines -> Information flows
o Solid lines -> Material flows (if information is delivered via paper it is also shown solid)
o Suppliers/Customers
o Own company

Question 10

Control Strategies in Production

Answer 10

• Push Control
• Pull Control
• Capacity Levelling

Question 11

Push Control

Answer 11

Based on specifications from production planning, products to be manufactured are pushed into production.
 High delivery capacity can be guaranteed through intermediate and final storage
 Optimal utilization of production machines and employees
 Disadvantages: Wastage in the form of stocks; Low Flexibility regarding customer’s change requests

Question 12

Pull Control

Answer 12

Production is only started when a customer order is received or stocks have reached an individually defined minimum.
 Reduces stocks as well as search and transport efforts
 Disadvantage: High dependency on suppliers

Question 13

Capacity Levelling

Answer 13

Smoothing of production orders
 Production orders are distributed evenly with the aim of evenly utilizing all capacities along the process chain

Question 14

Kaizen (LEAN Principle)

Answer 14

Generic Term for continuous improvement of a product or process

Question 15

SMED (LEAN Principle)

Answer 15

Single Minute Exchange of Die

Technique to reduce set-up time of production machines or lines

Question 16

FTT (LEAN Principle)

Answer 16

First Time Through

Percentage of parts which do not have to be reworked

Question 17

3PL Warehouse (LEAN Principle)

Answer 17

Third Party Logistics Warehouse

External Logistics service provider who also offers the storage of goods besides transportation

Question 18

Four Phase of business Analytics

Answer 18

1. Descriptive Analytics
2. Diagnostic Analytics
3. Predictive Analytics
4. Prescriptive Analytics

Question 19

Descriptive Analytics

Answer 19

o Key Question: What happened?
o Collection and analysis of (production) data with the goal of generation information
o Exclusive use of information from the past (historical data)
o Aggregation and targeted consolidation of existing data
o Clearly arranged presentation of information through tables, figures and graphics

-> Descriptive Analytics is used to understand current events in the production process and to describe different aspects in an aggregated way

-> Applied methods of data acquisition and processing:
* ETL (Extract, Transform, Load)  Data from different acquisition systems is combined in a data warehouse
* Data Warehouse  Cross-database serves as a data base for analytical processes in the following. Does not directly access the data of the operating business and thus does not interfere
* OLAP (Online Analytical Processing)  Provides multi-dimensional visual processing of data in data cubes.

Question 20

Diagnostic Analytics

Answer 20

o Key Question: Why did it happen?
o Detailed analysis of recorded data with the aim of identifying the cause
o Development of knowledge by recognizing patterns
o Recognition of connections and interdependencies or correlations

-> By the identification and the subsequent analysis of the event’s cause the user gets a deeper insight into the production process

-> The central concept of Diagnostic Analytics is Data Mining:
* Data evaluation with statistical methods to uncover new links, connections and trends
* Methods and objectives: Outlier detection, clustering, classification, association analysis, regression analysis, etc.
* Apriori algorithm (an example of association analysis)

Question 21

Apriori Algorithm

Answer 21

o Frequent use in the context of market basket analysis
o Using the a priori gene the frequency of purchase of goods is determined and afterwards the correlation
o Result of the algorithm is a correlation of the form “If shampoo and aftershave were bought, shaving foam was also bought in 90% of cases.

Question 22

Predictive Analytics

Answer 22

o Key Question: What will happen?
o The combination of existing data with rules and algorithms allows a prediction of prospective, probably occurring events
o Identification of opportunities and risks on the basis of the comparison of current and historical data
o Prediction of complex (economic) contexts
o Possibility of evaluation of potential fields of action
 Predictive Analytics is used in various fields of research nowadays, to predict behavior and future events in series production

-> Analytical Methods: For the generation of collected data numerous models and algorithms are suitable (e.g. Bayesian classification, Sequence Discovery, regression analysis etc.)
* Exponential smoothing
o

Question 23

Exponential Smoothing

Answer 23

Method of time series analysis for short-term forecasting from a sample with periodic historical data
o Mainly used when the time series reveals no systematic pattern
o Data with increasing actuality refers a higher weighting

Question 24

Prescriptive Analytics

Answer 24

o Key Question: What needs to be done to prevent it?
o Automatic synchronization of data, mathematical calculations and business rules lead to decision-making ability of enterprises
o Concrete forecasts about type, scope, timing and reason likelihood of occurring events
o Information about required actions with the goal of achieving predicted results
o Demonstration of the resulting effects based on any decision and the mutual influence

-> Using prescriptive analytics companies are able to anticipate the influence on their production and to optimize them before the occurrence of errors.

->Optimization methods
* Three main components of an optimization problem: Variables, the overall objective and constraints
* To describe different cases, suitable optimization models have to be selected
o Linear optimization (Cost – Quantity)
o Non-linear optimization (Price – Demand)
o Integer optimization (Bus – Passengers)