Life Cycle Analysis

Lecture 7 Impact Assessment

Course design credit: Scott Matthews, Chris Hendrickson, Deanna Matthews @ Carnegie
Mellon University
Multifunctional System
Typical Unit Processes
 For 1 product.. Emission(s)

Input(s) from
Product
Technosphere Unit Process +
Function

Inputs from Nature

 When there is only one product, there’s an obvious and direct connection
between the (1) product and the input and output flows
Multifunction systems
 May be multiple products / co-products
 More importantly, may have multiple functions
Allocation
 Attributing the input and output flows via some mathematical
relationship to the various products
 Try to avoid need for allocation!
 e.g., by breaking unit process down into smaller pieces
 Ex: Make more detailed model with specific sub-processes for making
ice (only input), making ice cream (only input). Problem ‘solved’ – but
more work
 Each of these add process steps, make it clearer what was connected
specifically to products
Allocation Steps (cont.)
2) If #1 fails, partition inputs/outputs to products based on “underlying
physical relationships”
e.g., mass, volume, etc. Allocate by percentage each

3) If #2 fails, use another method, e.g., by economic value (this is NOT a

default choice of method)
How much is A worth vs. B?
 Generally, should strive to use consistent allocation across system (not
‘picking and choosing’ for each)
 Special rules for recycling, etc.
Basic Allocation Example

 Imagine we were doing an LCA of pineapple

 Truck is a process to carry fruit (and baby) to market.
Inputs are fuel, etc. Outputs air emissions, etc.
 If we know truck used 2 gals. gasoline, only visible fruit
 How much gas allocated to each unit of 2 fruit types?
 Allocation Data – Small Group Work
Unallocated flow is 2 gallons of gasoline

Metric Pineapple Watermelon

Number of items 4 12

Mass 2 pounds each 5 pounds each

Economic Value $2 per pound $0.80 per pound

Don’t forget to validate that sum of allocated effect across all = 1 !

This example excludes the baby!

Modeling the End of Life
Waste treatment, recycling, and reuse
End of Life Waste Treatment
 Theoretical Case  OpenLCA Practice
 Use phase
Input Example Input Example

Product Input A plastic bag Product Input A plastic bag

Output
Output
Product Output A used bag
Waste output A used bag Waste A waste bag

Note: In OpenLCA, we cannot use waste flow as Find a provider to treatment waste or create a
the reference output (product flow only). process to treatment waste;
Note: When we create a process for a waste flow,
the waste flow is listed in the input.
Reuse in the Product System
 A simple approach is to find the product flows that are right reusable
materials for your product system and include them and the
corresponding processes in the correct process of your product system.
 In case that you cannot find these through database (and it is likely to
happen), then you need to create a process that produces the reusable
materials first. Other inputs
 Cut-off:
Used Products
Manufacture Use Recycling

Recycled materials
Data Methods for LCA
Problem of Unknown Numbers
▪ If we need a piece of data, we can:
▪ Look it up in a reference source
▪ Collect number through survey/investigation
▪ Guess it ourselves
▪ Get experts to help you guess it
▪ Often only ‘ballpark’, ‘back of the envelope’ or ‘order of magnitude
needed
▪ Situations when actual number is unavailable or where rough estimates are good
enough
▪ e.g., 100s, 1000s, … (102, 103, etc.)

▪ Source: Mosteller handout

In the absence of “Real Data”
▪ Are there similar or related values that we know or can guess?
(proxies)
▪ Mosteller: registered voters and population
▪ Are there ‘rules of thumb’ in the area?
▪ E.g. ‘Rule of 72’ for compound interest
▪ r*t = 72: investment at 6% doubles in 12 yrs
▪ MEANS construction manual
▪ Set up a ‘model’ to estimate the unknown
▪ Linear, product, etc. functional forms
▪ Divide and conquer
Primary and Secondary Sources
 Primary Source:
A reference, article, etc., that is the original source of data or results.
 Example: US government agency - data on inflation, $ output,
unemployment, etc.
 Secondary Source:
A reference that repeats the data/results found in a primary source
 Example: An article in The New York Times discussing last month’s unemployment
(9%) which likely came from the Department of Labor
Attributes of Good Assumptions
▪ Need to document assumptions in course
▪ Write them out and cite your sources
▪ Have some basis in known facts or experience
▪ Write why you make the specific assumptions
▪ Are unbiased towards the answer
▪ Example: what is inflation rate next year?
▪ Is past inflation a good predictor?
▪ Can I find current inflation?
▪ Should I assume change from current conditions?
▪ We typically use history to guide us
Environmental Impacts
LCA was first developed to evaluate environmental impact!
Impacts
Effects of Interest
 Our inspiration for wanting to study a life cycle may be varied.
 Economics

 Energy

 Airemissions
 Water emissions

 Human labor issues

 For any of these items of concern, we would need to take our structure
and then find useful data
Indicators to Impacts
 Really, we probably are motivated by impacts not just indicators.

 Energy use -> fossil fuel depletion

 Greenhouse gas emissions -> global warming
 CFCs -> ozone depletion
Impact Assessment
 Involves estimating significance of environmental impacts studied
 e.g.,
Not just GHG emissions, but global warming
 Need frameworks to “convert” emissions into final impacts

 Least “mature” portion of LCA

 Still
subject of ongoing science and research
 Can be subjective given possible choices and values needed

 High uncertainty
 Impact Assessment
24

 Impact assessment not new. LCA did not invent it. Scientists doing it for
decades.
 Environmentalimpact assessment, risk assessment, performance
benchmarking, etc.
 Key feature of LCIA vs. other frameworks is link to a particular
functional unit (and of course the entire life cycle as a boundary),
 Focuses our attention on impacts as a function of that specific normalized
quantity.
LCIA Methods Used
 A bit too much to discuss now, but generally, this means listing the
specific impact assessment methods you will use
 For example, if you care about climate change, you’ll use some sort of
method from IPCC or others that deals with that.
Impact Assessment
in the ISO LCA Framework (LCIA)
 Mandatory Elements:
 Selection (impact categories, their indicators, and characterization models)
 Classification (assigning LCI results to categories)
 Characterization (calculation of category results)
 i.e., - at least get things into correct categories

 Optional Elements:
 Normalization (comparing to reference info)
 Grouping (sorting/ranking impact categories)
 Weighting (with numerical factors/value choices)
 Data Quality Analysis (uncertainty/sensitivity)
Cause-Effect Chain
Specific to an emissions example
 Cause-Effect Chain
Specific to an emissions example

Methyl bromide release

from pesticide use
Similar chains could be drawn for
resource depletion or land use – try it!

CH3Br mg/m3

Increased Loss of
radiative forcing atmospheric ozone

Rising sea level

Coral reef damage Increased skin
Increased severe weather cancer
Midpoints vs. Endpoints
 Typical LCI result is ‘an emission, waste generation, etc.’ - leads to a
midpoint

 Breathing emissions, etc. potentially leads to health effects and

damages - these are endpoints
 This can get pretty complicated

 Now you can see why many studies stop at LCI stage (not full LCA)!
Characterization
 Transforms classified flows into impact category indicators via
characterization factors
 Impact category indicators relevant to resources, ecosystems, human health
 These condense multiple units into a single common unit

 Characterization factors are the result of separate scientific studies on

impact assessment (not done for a single LCA study)
Characterization (cont.)
 GHGs - obvious example. Use GWP weights! (e.g., CO2-equivalents
or CO2e)
 CO2 : 1 kg of CO2 = 1 kg of CO2e
 CH4 : 1 kg of CO2 = 28 kg of CO2e
 N2O : 1 kg of CO2 = 265 kg of CO2e
 Can then sum up individual effects into a total (units CO2e)

 For many studies, the only LCIA done is global warming (thus the
“carbon footprint” tag)
Optional Step - Normalization
 Dividing by a selected reference value
 “Normalizes” against some baseline, e.g., total effect of a person per
year
 Different normalization basis for each impact (e.g., total in a county/region,
total per-capita)
 Rationale: Global warming impact might be 50 tons. How important is that?

 Note: can also normalize against one of the other options being studied (A
vs B)
 Downside – normalizing based on total effects generally yields
negligible values
Normalization (2)
 The Standard does not provide these values
 But others have generated values that often used (and are integrated into
software)
Normalization factor availability
 US
 Canada
 Many European countries
 Global for a few impact categories
 Normalized Results
37

 Building on example..
 Sample Output
38

 What are useful, problematic issues?

Interpretation
 How do results fit goal/scope?
 Need to “answer questions” of goal and scope
 Assessment of data quality
 Sensitivity analysis on inputs/outputs
 Conclusions and recommendations
 .. and hopefully Improvement
 Ex:
Are any parts of the inventory obvious targets for re-design?
 Material with high energy requirements, Life cycle stage that dominates
Reporting and Optional Review
 Report
 Critical part (and why we spend time on it)
 Is the record of the study, and an indicator of the study quality itself

 May or may not be made public (most are not)

 Peer Review
 Required if making comparative assertions for public release (not required
if just public)
 Checking for “ISO compliance”

 Chance for inconsistencies, errors, etc. found

 A growing practice (we do lots of them)

Phases of an LCA* - Iterative

 Not a once through process - all phases are

iterative!

 Adjust as you go along

 Ex 1: trying to find data for a process, can’t find
it, redo the scope/boundary!
 Ex 2: realize you can’t make a solid conclusion
given results, need to refine scope or collect more
data in inventory stage
 Ex 3: unexpected result. May need to add an
impact or category

 Changes via iteration happen in ~100% of

studies. It is very common.
43 Life Cycle Costing
 Objectives
44

 Choose amongst alternatives for cheapest cost over life cycle

 Connect idea of life cycle costing with environmental life cycle thinking
Point of Departure:
Life Cycle Cost Analysis (LCCA)

 Developed to track costs over life cycle of infrastructure, etc.

 Ensure that decision makers thought about “first cost” and maintenance,
etc.
 Useful also for personal decisions (e.g., what kind of car to buy)
 If you have taken an investment analysis course, this is familiar
 In depth stuff on this topic better left for a course on Infrastructure Management
 Other domains – total cost of ownership
Objectives of LCCA
 Identify costs that happen during the life of the “project” that are not
first cost
 Determine a “total cost”
 Compare these to initial (“first”) cost
 Ideally, “discount” all future costs to the present for comparison
 Ideally, compare to total benefits to determine whether it is worthwhile
Example - Light Bulbs
 Should I choose incandescent or fluorescent or LED?
 Bulbcosts $1 versus $3 versus $10-20)
 How long are they on? Hours -> kWh -> $
 How long do they last? 1000 vs 8000 hrs
 How much do they “cost to buy & change”
◼ e.g., campus laborers cost $40/hr
 Fluorescent bulbs can save lots over life.
 Life Cycle Costs of a Passenger Car
 What are life cycle costs for a
car?
 Edmunds.com – “True cost to
own” calculator
 http://www.edmunds.com/tco.html
 Which is cheaper? Why? What
might change our decision?
 Is it affected by the type of vehicle
(e.g., electric versus hybrid)?
Social Life Cycle Analysis
Social LCA
 Social LCA usually refers to the assessment of the real and potential
social and socio-economic impacts of goods or services including
positive and negative impacts along their life cycle.
 Environmental LCA impacts are mostly negative
 Social impacts may also be distinctly positive
The framework of SLCA

Source: S, Sala., et al., 2015

Social Hotspots Database
 The SHDB provides great insight into social hotspots in product supply
chains, covering 140 countries and regions and 57 economic sectors.
The database includes an extensive list of indicators around labor
rights, health and safety, human rights, governance, and community
infrastructure.
Summary
Criticisms / Limitations of LCA
 No single, accepted method
 Data reliability and quality is questionable.
 Models based on assumptions.
 Problem boundaries are arbitrary.
 Scale issues - global -> local, etc.
 Uncertainty is everywhere
 Spatial and temporal issues
 Comparisons between studies difficult
Alternative Modeling Approaches
 Risk analysis
 Environmental impact assessment
 Environmental Audit
 Benefit-cost analysis

 Life cycle impact assessment is not required to address economic or

social aspects (key part of sustainability!) and so may differ from
above.
 Now there are more efforts to include economic and social aspects
57