2 Modelling a Product Usage Scenario
2.1 Source of Data Collection
In order to model a usage scenario, it is necessary to gather data from the use phase.
Different kinds of data can be collected in order to build usage scenarios. In this
paper, four sources of data are identified: real-time data, experimental data,
numerical simulation data and average scenario data. They are analysed based on
four criteria: (i) time needed to collect data, (ii) cost to collect data, (iii) reliability of
data and (iv) accuracy of real usage data (Table 1).
Real-time data are collected directly with the help on sensors while the product is
running. It is the most reliable kind of data and allows the analysis of the actual
usage specific to each user. However, data collection and analysis can be
time-consuming and/or costly, as it is necessary to use multiple sensors and allocate
resources to analyse data. In addition, it implies that the environmental footprint
must be evaluated continuously or only at the end of the product life. The environmental footprint for a specific set of data will be accurate for a specific usage but
rather specific to it, i.e. not necessarily pertinent for another. It can be helpful to
build an average scenario, specific to each type of business (e.g. fresh or frozen
product, international long haul, national long haul delivery or urban delivery).
Experimental data are collected during punctual measurement on the product in
order to recreate real condition of use. Again, collection can be time-consuming and
costly, depending on the resources used. However, it requires determining precise
parameters to analyze in order to have reliable data. It does allow for analysing the
influence of single parameters or for mixing the influence of different parameters. It
does allow for creating specific scenarios that can be accurate with real usage.
Data from numerical simulations are based on mathematical routines recreating
real-world use conditions. The approach to data collection requires a thorough
knowledge of the different parameters to set up the simulation and interpret the
results. In some cases, especially with products requiring the integration with
another system, it can be difficult to control all the external parameters influencing
Table 1 Proposal of quotation of four sources of data to evaluate the environmental performance
of the use phase
a
#
Solution
Time
needed
Cost
Reliability of
data
Accuracy of real
usage data
1
Real-time data
−−
−
++
++
2
Experimental data
−
−
+
+
3
Numerical
simulation data
+
+
−
+
4
Average scenario
data
++
++
−−
−−
a Quotation ++ very good; + good; −bad; −− very bad
Is It Useful to Improve Modelling of Usage Scenarios …
233
2.1 Source of Data Collection
In order to model a usage scenario, it is necessary to gather data from the use phase.
Different kinds of data can be collected in order to build usage scenarios. In this
paper, four sources of data are identified: real-time data, experimental data,
numerical simulation data and average scenario data. They are analysed based on
four criteria: (i) time needed to collect data, (ii) cost to collect data, (iii) reliability of
data and (iv) accuracy of real usage data (Table 1).
Real-time data are collected directly with the help on sensors while the product is
running. It is the most reliable kind of data and allows the analysis of the actual
usage specific to each user. However, data collection and analysis can be
time-consuming and/or costly, as it is necessary to use multiple sensors and allocate
resources to analyse data. In addition, it implies that the environmental footprint
must be evaluated continuously or only at the end of the product life. The environmental footprint for a specific set of data will be accurate for a specific usage but
rather specific to it, i.e. not necessarily pertinent for another. It can be helpful to
build an average scenario, specific to each type of business (e.g. fresh or frozen
product, international long haul, national long haul delivery or urban delivery).
Experimental data are collected during punctual measurement on the product in
order to recreate real condition of use. Again, collection can be time-consuming and
costly, depending on the resources used. However, it requires determining precise
parameters to analyze in order to have reliable data. It does allow for analysing the
influence of single parameters or for mixing the influence of different parameters. It
does allow for creating specific scenarios that can be accurate with real usage.
Data from numerical simulations are based on mathematical routines recreating
real-world use conditions. The approach to data collection requires a thorough
knowledge of the different parameters to set up the simulation and interpret the
results. In some cases, especially with products requiring the integration with
another system, it can be difficult to control all the external parameters influencing
Table 1 Proposal of quotation of four sources of data to evaluate the environmental performance
of the use phase
a
#
Solution
Time
needed
Cost
Reliability of
data
Accuracy of real
usage data
1
Real-time data
−−
−
++
++
2
Experimental data
−
−
+
+
3
Numerical
simulation data
+
+
−
+
4
Average scenario
data
++
++
−−
−−
a Quotation ++ very good; + good; −bad; −− very bad
Is It Useful to Improve Modelling of Usage Scenarios …
233