Page 296 - DMGT523_LOGISTICS_AND_SUPPLY_CHAIN_MANAGEMENT

Unit 13: Logistics Design and Operational Planning

Optimization uses linear or mathematical programming to evaluate alternatives and select the Notes
best one. While it has the benefit of being able to select the best option, optimization applications
are often smaller in scope than typical simulation approaches. Because of its powerful capabilities,
optimization is used extensively for evaluating logistics network alternatives such as the number
and location of distribution centres.
Defining and Reviewing Assumptions

Assumption definition and review builds on the situation analysis, project objectives, constraints,
and measurement standards. For planning purposes, the assumptions define the key operating
characteristics, variables, and economics of current and alternative systems. While the format
will differ by project, assumptions generally fall into three classes: (1) business assumptions,
(2) management assumptions, and (3) analysis assumptions.

Business assumptions define the characteristics of the general business environment, including
relevant market, consumer, and product trends and competitive actions. The assumptions define
the broad environment within which an alternative logistics plan must operate. Business
assumptions are generally outside the ability of the firm to change.
Management assumptions define the physical and economic characteristics of the current or
alternative logistics environment and are generally within the firm’s ability to change or refine.
Typical management assumptions include a definition of alternative distribution facilities,
transport modes, logistics processes, and fixed and variable cost.
Analysis assumptions define the constraints and limitations that must be included to fit the
problem to the analysis technique. These assumptions frequently concern problem size, degree
of analysis detail, and solution methodology.

Identifying Data Sources

In actual practice, the process of data collection begins with a feasibility assessment. In addition,
a fairly detailed specification of data is required to formulate or fit the analytical technique.
However, at this point in the planning procedure, detailed data must be collected and organized
to support the analysis. For situations when data is extremely difficult to collect or when the
necessary level of accuracy is unknown, sensitivity analysis can be used to identify data collection
requirements.

Example: An initial analysis may be completed using transportation costs estimated
with distance-based regressions. If analysis indicates that the best answer is very sensitive to the
actual freight rates, there should be additional effort to obtain more precise transport rates from
carrier quotes.

Once operational, sensitivity analysis can be used to determine the major factors involved.
When these factors, such as outbound transportation expense, are identified, more effort can be
directed to increasing transportation accuracy; correspondingly, less effort can be directed toward
other data requirements.
The majority of data required in a logistical study can be obtained from internal records. Although
considerable searching may be needed, most information is generally available.
The first major data category is sales and customer orders. The annual sales forecast and percentage
of sales by month, as well as seasonality patterns, are usually necessary to determine logistics
volume and activity levels. Historical samples of customer invoices are also necessary to determine
shipping patterns by market and shipment size. The combination of aggregate measures of demand
and shipment profiles characterizes the logistics requirements that must be met.

LOVELY PROFESSIONAL UNIVERSITY 291