Go to the CCL Discount Scheme homepage

 

  

Developing the Product Mix and Output Algorithm

Nine Common Approaches

Introduction

The approaches have been developed to help demonstrate how the PMOA can be applied in the most common situations.

It is not mandatory to use one of the common approaches (CA’s), however, by using one or more of the CA’s, it should be:

      • easier for you to prepare a submission to DEFRA;
      • easier for DEFRA to review your submission;
      • more likely that your submission will be approved quickly;
      • easier for the discount scheme to help.

The CA’s developed by the discount scheme are:

Common Approach 1

Single Product, Single Regression

Common Approach 2

Mixed Product, Multiple Regression

Common Approach 3

Mixed Product, Non-Coinciding

Common Approach 4

Mixed Product, Complex Production

Common Approach 5

Specific Event, with Sub-Meters

Common Approach 6

Specific Event, without Sub-Meters

Common Approach 7

Introduction of a New Product

Common Approach 8

Discontinuation of a Product

Common Approach 9

Incorporating Non-production Related Variables, e.g. degree days

Together with the methodology described for each CA, dummy submission documents have been prepared to demonstrate how each CA can be used in your pre-approval document.

The aim of the PMOA is to allow you to adjust your target to take into account a change of circumstances. The methodology described for each common approach, describes how your target should be adjusted using the data for your milestone year, e.g. energy use and production data. The pre-approval documents should demonstrate how you propose to adjust your target when your milestone year data becomes available.

This document should be read in conjunction with "Applying the PMOA – Two Simple Examples" and the dummy submission documents.

Common Approach 1 – Single Product, Single Regression

General Characteristics:

Only one product made at the site.

Output has decreased since the base year.

No other influencing factors (such as weather for example).

Data Requirements:

Main utility meters and production information:

  • annual data for each utility and total production for the Base Year and Target Year;

  • regular data points (e.g. weekly) collected during either the base year or during a more recent period. Need a minimum of 15 data point

Methodology:

For regular data points obtained during the base year, see PMOA Simple Example 1A.

For regular data points obtained during a more recent period, see PMOA Simple Example 1B.

Possible Problems:

If you obtain a poor correlation it could be because there is really more than one product type, or there is a non-production related variable which influences energy use and should be taken into account (e.g. weather).

Check the consistency of your data, has the energy and production data been collected for exactly the same period, have any readings been miss-read?

Common Approach 2 – Mixed Product, Multiple Regression

General Characteristics:

More than one product type or influencing variable

  • but not too many
Data Requirements:

Main utility meters and production information:

  • annual production data for each production type and annual data for each energy source for both the Base Year and Target Year;

  • regular data points (e.g. weekly) collected during either the base year or during a more recent period. Need a minimum of 25 data point
Methodology:

For regular data points obtained during the base year, see PMOA Simple Example 2A.

For regular data points obtained during a more recent period, see PMOA Simple Example 2B
Possible Problems:

If you obtain a poor correlation it could be because there are too many or too few product types, or there is a non-production related variable which influences energy use that has not been taken into account (e.g. weather).

Check the consistency of your data, has the energy and production data been collected for exactly the same period, have any readings been miss-read?

Consider another approach if still achieve poor correlation

Common Approach 3 – Mixed Product, Non-Coinciding

General Characteristics:

Two or more products

Only one product made at any given time

Usually, same equipment used for all product types

Two types of production cycle:

long cycle variations:
e.g. seasonal products like frozen vegetables, weekly data can be split into groups for each product type

short cycle variations:
e.g. different recipes made for a few hours or days, need to set-up data collection at a suitably short interval, may be best to do a short data collection campaign to evaluate energy and production relationship

Data Requirements:

Main utility meters and production information:

  • annual production data for each production type and annual data for each energy source for both the Base Year and Target Year
  • regular data points (at a frequency suitable to the production cycle) collected during either the base year or during a more recent period when all products are being manufactured
  • data outlining production changes, e.g. dates/times of production of each product

Methodology:

Identify the energy use of each product and undertake a linear regression as shown in Simple Example 1 for each product type
Possible Problems: Data collection accuracy, non-production related influencing variables should be incorporated (in this case a multiple regression analysis should be done for this product type instead of a linear regression)

Common Approach 4 – Mixed Product, Complex Production

General Characteristics:

Numerous product types

Common Approach 2 gives poor correlation

Data Requirements:

Assessment of energy use of different production lines is essential.

Use sub-meters if available.

Where sub-meters are not available other forms of analysis will be required, e.g. theoretical energy use

Methodology:

Regression analysis of sub-metered data (Simple Example 2).

Theoretical calculations based on product specifications.

Corrections for operating hours.

Corrections for degree days.

Spot check data.
Possible Problems: Data collection accuracy, correct statistical analysis of data. Correct definition of variables.

Please note that no dummy submission document has been prepared for this common approach since all sites following this methodology will be in different situations.

Common Approach 5 – Specific Event, with Sub-Meters

General Characteristics:

A one-off change that occurs to the process

  • e.g. an extra processing step is added

Many regulatory/relevant constraints can be demonstrated following this approach

Data Requirements:

Main utility meters and production information:

  • annual production data for production and energy for both the Base Year and Target Year;

  • data from the sub-meter(s) on the new process to demonstrate extra energy use;

  • data outlining changes, e.g. dates/times of installation of new proces

 

Methodology:

Calculate adjusted base year SEC as if new processing step had been operating, e.g:

SEC adjusted base year = SEC base year + SEC new process

Calculate the adjusted first milestone target using the adjusted base year SEC
Possible Problems: Data collection accuracy, availability of sub-metered data. Correct definition of variables

Common Approach 6 – Specific Event, without Sub-Meters

General Characteristics: Identical to Common Approach 5 except without sub-meters
Data Requirements:

Main utility meters and production information:

  • annual production data for production and energy for both the Base Year and Target Year;

  • theoretical calculations to support data analysis;

  • data outlining changes, e.g. dates/times of installation of new process
Methodology:

Carry out a regression analysis of energy and production data before and the process change to identify the impact of the step change.

Use CUSUM technique if appropriate (see Good Practice Guide 112 from the Energy Efficiency Best Practice Program).

Calculate adjusted base year SEC and adjusted milestone target

.
Possible Problems: Data collection accuracy and statistical analysis. Correct definition of variables

Common Approach 7 – Introduction of a New Product

General Characteristics:

New product has been brought on-line since the base year

Data Requirements:

Main utility meters and production information:

  • annual production data for each production type and annual data for each energy source for both the Base Year and Target Year;

  • regular data points (e.g. weekly) collected during a recent period. Regular data collected for the base year is not applicable since the new product was not made during the base year. Need a minimum of 25 data point

 

Methodology:

Data collection and analysis as PMOA Simple Example 2B, except the production value of the new product for the base year will be zero
Possible Problems: Data collection accuracy and statistical analysis. Correct definition of variables

Common Approach 8 – Discontinuation of a Product

General Characteristics:

A product has stopped being manufactured since the base year

Data Requirements:

Main utility meters and production information:

  • annual production data for each production type and annual data for each energy source for both the Base Year and Target Year;

  • regular data points (e.g. monthly or weekly) collected during the base year. Regular data collected for a recent period is not applicable since the discontinued product has not been made during this period.

  • theoretical calculations to support data analysis

Methodology:

Data collection and analysis as PMOA Simple Example 2A on base year data. Since accuracy and frequency of this data may be poor, use theoretical calculations to support energy consumption of the discontinued product
Possible Problems: Availability of monthly or weekly base year data, or information to support theoretical calculations

Common Approach 9 – Incorporating Non-Production Related Variables, e.g. degree days

General Characteristics:

Factors other than production significantly influence energy use

Data Requirements:

Main utility meters and production information:

  • annual production data for each production type and annual data for each energy source for both the Base Year and Target Year;

  • regular data points (e.g. weekly) collected during the base year or a recent period. Need a minimum of 25 data points

Data for the non-production related variable:

  • an annual value for the non-production related variable for the Base Year and Target Year;

  • regular data points gathered at the same interval and period as the energy and production data

Examples of non-production related variables include:

Methodology:

Data collection and analysis as PMOA Simple Example 2A or 2B depending upon when the regular data was collected. The influencing variable is treated in the same way as a production value in the analysis
Possible Problems: Data collection accuracy and statistical analysis. Correct definition of production and non-production variables

 

.

.


 
 
[CCL Home]