Enterprise Performance and ERP

1

• Enterprise Performance and ERP
• Dr. Con Sheahan

University of Limerick
June 14th 2005
Department of Manufacturing and Operations
Engineering
2
What Does the Scheduling Practitioner Really
Want?

• Does the concept of Optimality appeal to a
  practitioner?
• The practitioner is very pleased with a Feasible
  Solution.
• The practitioner is sometimes pleased with a
  declaration that there is no Feasible Solution.
• It confirms his own empirical view so he can go
  in to the production meeting with an evidence
  based request for the required extra resources.

3
What Does the Scheduling Practitioner Really
Want?

• A typical response of the practitioner is
• I am not worried about the jobs that can be
  scheduled, I am concerned about the jobs that
  cannot be scheduled.
• What are the Jobs That can hurt me?
• My Translation
• The practitioner is firstly concerned with jobs
  that have resource constraint conflicts.
• The practitioner wants the Diagnostic process of
  a proposed schedule to be focused on identifying
  those jobs that are going to most negatively
  effect his Performance Outcomes.

4
Opportunities and ChallengesPractitioner
Perspective

• How does the Scheduler change the lived
  experience of the Practitioners Involved?
• Is this schedule the Best?
• How does the practitioner verify and gain
  credibility and confidence in the outputs of the
  scheduling tool?
• Does the scheduler enable the practitioner to
  answer the myriad of order promise choices? e.g.
• Can this order be made by a certain date?
• If this work order is delayed by x days will the
  associated sales order go overdue?
• What is the economic consequences of the
  increased Machine Utilisation Vs On-time
  delivery, what if 97?

5
Performance ManagementThe Balanced
Scorecard(Source Kaplan Norton, 1996)

• 1. Financial How do we look to our Shareholders?
• 2. Customer How do our Customers See Us?
• 3. Internal Business Process What should we do
  that is Excellent?
• 4. Employee and Organization Innovation and
  Learning Can we continue to Improve and Add
  Value?
• WhiteboardWhat KPIs are being used in practice?

Balanced Scorecard Perspectives
6
Performance MeasurementCurrent Practice

• Many of the commercial applications currently on
  offer focus on the need to integrate the data
  from a wide range of information generating
  systems such as ERP, CRM and legacy applications.
• Once the Performance Metric is calculated it is
  the responsibility of the decision maker to
  extract the meaning from this particular output
  with a reference to data warehouses for a more
  detailed analysis of the raw data.
• A great deal of focus is on the Visual Indicators
  e.g. Dashboards. Perhaps a richer visual
  experience is needed however the timing of this
  information is problematic?

7
Enterprise Performance Management

• Enterprise Performance Management within a
  manufacturing context aims to arrive at the most
  effective solution to enterprise problems by the
  optimum utilisation of
• A set of resources (time, material, people,
  machines and money etc.)
• with respect to
• A portfolio of performance metrics.

8
Enterprise Performance Management

• When you check the internet search engines
  however these terms are typically applied to
  computer applications and network performance
  with the enterprise rather than the performance
  of the enterprise itself.
• When the ERP vendor web sites are searched, there
  appears to be little focus on software modules
  that specifically support explicit performance
  outcomes.
• There is a myriad of reports on what went wrong
  in the past via data mining but very few holistic
  analyses.

9
Enterprise Performance Metric Portfolio

• A good place to start is the metric set proposed
  by Goldratt
• Throughput (T),
• Inventory (I) and
• Operating Expense (OE)
• If we add to this On time in Full (OTIF) type
  metric for delivery.
• It would be good to know where my Bottlenecks
  are.
• One then has a useful set of performance metrics
  that are find reasonable support in the
  literature.

10
MRP Inputs Outputs (From Slack)
Master Production Schedule
Material Requirements Planning
Product Structure File
Inventory Master File
Planned Order Releases
Work Orders
Purchase Orders
Rescheduling Notices
11
Capacity Requirements Planning
12
MRP II Flowchart
Business Plan
Marketing Plan
Financial Plan
Production Plan
Feasible?
No
more
Yes
13
Feedback
14
Master Production Schedule (MPS)

• A firms Performance aspirations are typically
  initially expressed via the MPS process.
• How does on verify that this MPS instance enables
  one to accomplish the performance metric
  portfolio that has been advocated for the
  enterprise.
• One finds this out only after the MPS has been
  executed???

15
MRP Outputs

• A typical MRP application execution generates a
  large number of action messages.
• Many practitioners are swamped by a large number
  of messages with an nil defined effect on the
  portfolio of performance outcomes.
• Is this why despite considerable ERP investments
  there still tends to be a great deal of reactive
  response to crises (fire fighting) to predictable
  production planning situations.
• It is a case of where the important signals are
  hidden within the noise of the unimportant
  signals.

16
Bottleneck Identification

• The identification of bottlenecks and their
  consequences is a important work load driver for
  the manufacturing decision maker.
• The location and consequences of a bottleneck can
  be determined manually from the data provided by
  the ERP Infrastructure.
• When the ERP or Enterprise Resource Planning is
  used what is actually meant?
• How much is it assisting the practitioner?
• If it is not contributing on bottlenecks but
  leaving it to the decision maker hence the
  popularity of spreadsheets for slicing and dicing
  ERP data in order to turn it into information.

17
Problem Solving

• Once a Root Cause of why a Sales Order is not
  going to achieve its expected performance
  criterion.
• E.g. On Time in Full
• Is it Capacity?
• Is it Tooling?
• Is it Quality of components?
• Are the skilled people available?. E.g. Set-ups.

18
Diagnosis of Performance Metrics

• An important issue for most practitioners is the
  diagnostic capability of an enterprise
  performance management system.
• e.g. Is it possible to assign a root cause to a
  predicted sales order overdue?
• Will it be delayed in manufacturing?
• Will it be delayed in material procurement?
• Will it be delayed due to a long queue in quality
  inspection?
• In all of these cases the remedy is quite
  different.

19
Enterprise Performance Management Implementation
Methodology

• The identification of the appropriate performance
  optimisation criteria at the Enterprise,
  Division, Plant, Manufacturing Cell and
  Facility/Machine , Person level.
• The provision of the appropriate high quality
  information to complete the performance
  optimisation.
• Develop the Performance Optimisation Algorithms
  to yield the Best Schedule.

20
Concept Challenges

• Data Acquisition and Integration.
• Performance Metric Portfolio Definition
• Enterprise Modelling (Manufacturing Process and
  Resources, Shifts and Calendars )
• Efficient Optimisation system generating
  near-optimal solutions.

21
Trade-Off Performance Metrics Goal Congruence

• In practice this is mostly frequently done on a
  subjective basis. Marketing insists that the
  order must be shipped for strategic reasons.
• e.g. When it is possible to obtain increased
  Material Utilisation, is it acceptable to delay a
  customer order?
• Simply providing the trade-off cost in real time
  may be sufficient.
• Procurement is measured on Inventory levels.
• Manufacturing is measured on Resource
  Utilisation.
• Sales are measured on customer service.

22
Exploitation of Performance Metrics

• A challenge to exploiting Performance Metrics in
  this form is that they are very difficult to
  transform into meaningful objective functions
  within optimisation applications.
• A practical limitation of many metrics is that
  the original promise or commitment is not
  feasible. (e.g. Promising 20 of incoming orders
  to be delivered in 8 weeks when there is a 12
  week backlog.)
• Is the Performance Metric amenable to improvement
  with optimisation models. e.g. Quality Failure
  rates.

23
Conclusions

• ERP is a reasonable data infrastructure.
• Many applications have a considerable Total Cost
  of Ownership (TCO) however.
• While ERP vendor claims are quite strong in terms
  of making firms more profitable the modules they
  provide for Enterprise Performance Management are
  very modest if they exist at all.
• The concept of Enterprise Performance Management
  is relevant if practitioners can express their
  needs objectively.

24
Reality Check

• Is this attainable in a production situation?
• Is the Information available?
• How good does it have to be to be useful?
• Is the use of a suite of performance metrics the
  appropriate instrument to enable the practitioner
  to exploit the methodologies proposed by the
  theory proposed by the Planning, Scheduling and
  Control in Manufacturing?

25
Parting Thought

• In the absence of Enterprise Performance
  Management are two types of practitioner
  personae.
• There are those that are
• making good
• and those that are making good excuses!