Modeling All-Electric Blow Molding Payback in a Few Lines of Python

Plant engineers are often asked a deceptively simple question: "If we buy the all-electric blow molding machine instead of the hydraulic or hybrid one, when do we get the extra money back?" The arithmetic is easy. The hard part is feeding it inputs that describe the same product, measured the same way. This post turns that problem into a small, reusable model you can drop into a notebook or spreadsheet.

The logic follows a payback guide published by Suzhou Tongda Machinery, an extrusion blow molding machine maker founded in 1999 in Zhangjiagang that supplies customers in more than 100 countries. Their article on calculating all-electric blow molding ROI explains the reasoning in plant terms; here we restate it as code and a list of input rules.

The core idea

Simple payback is just:

payback_years = extra_installed_investment / annual_net_operating_savings

Everything interesting lives in the denominator, and every term in it should be normalised to accepted bottles, the ones that pass inspection.

Input rules before you write any code

1. Same product on both sides. Same bottle drawing, resin, target weight, mold, annual accepted quantity and schedule.
2. Metered energy, not nameplate power. Use kWh per 1,000 accepted bottles from comparable runs. Power ratings are for sizing the electrical supply.
3. State the meter boundary. Machine only, or machine plus chillers, compressors, granulators and inspection? A smaller reading on the machine meter may not translate to the whole line.
4. Keep weight tolerance and average weight apart. A tighter tolerance does not reduce average resin use unless a lower mean is actually demonstrated.
5. Count each improvement once. If annual accepted output already reflects rejects and stops, do not also book a separate downtime saving for the same effect.

A minimal model

def energy_saving(base_kwh_per_k, new_kwh_per_k, bottles, price_per_kwh):
    # kWh figures are per 1,000 accepted bottles
    delta_kwh = (base_kwh_per_k - new_kwh_per_k) * bottles / 1000
    return delta_kwh * price_per_kwh

def resin_saving_tonnes(mean_reduction_g, bottles):
    # only use a measured reduction in MEAN weight
    return mean_reduction_g * bottles / 1_000_000

def simple_payback(extra_investment, annual_net_saving):
    return extra_investment / annual_net_saving

Demand charges, tax, financing and residual value are deliberately left out. Simple payback is a screening number, not a full investment appraisal; it ignores the time value of money.

Plugging in the published example

Tongda's guide uses a hypothetical set of inputs to show the arithmetic: 8,000,000 accepted bottles a year, 25 versus 18 kWh per 1,000 bottles, USD 0.15 per kWh, USD 1,600 of other net savings and USD 30,000 of extra investment. Running those through the functions:

e = energy_saving(25, 18, 8_000_000, 0.15)   # 8400.0
net = e + 1600                               # 10000.0
simple_payback(30_000, net)                  # 3.0 years

The guide is explicit that these are illustrative inputs, not a quotation and not a measured result from any specific machine. That matters: the "three years" is a property of the inputs, not of all-electric drives in general.

Add a sensitivity loop

Utilisation is the lever most people underestimate. Hold everything else constant and vary volume:

for bottles in (8_000_000, 6_000_000, 4_000_000):
    net = energy_saving(25, 18, bottles, 0.15) + 1600
    print(bottles, round(net), round(simple_payback(30_000, net), 2))
# 8000000 10000 3.0
# 6000000 7900 3.8
# 4000000 5800 5.17

These match the sensitivity table in the original guide. In a real model, revisit which maintenance and operating costs scale with utilisation instead of holding them fixed.

Resin as a separate scenario

The guide also shows that, at 8 million bottles, a demonstrated 1 g reduction in mean bottle weight equals 8 tonnes of bottle material per year (resin_saving_tonnes(1, 8_000_000)). Keep this out of the energy figure unless you have measured it, and remember that flash, rejects, regrind and losses all affect actual resin purchasing.

What to ask a supplier for

To fill the model with real numbers, request accepted output, metered kWh with the boundary stated, average bottle weight and its variation, reject rates, utility conditions, annual operating hours and the full installed-cost scope. Keep theoretical ratings in a separate column from test results.

If you are evaluating equipment, Tongda's fully electric blow molding machine series page lists the current HSE models, and the company can be reached through its official website, tongdamachine.net.