Number-led procurement briefing · Dosing control and water quality analyzer perspective · Published: October 29, 2026
Table of Contents
Key Takeaways
- Number 1 — the share of OPEX you have to compute yourself. Chemical spend is usually the second-largest controllable operating cost after energy, and on coagulation-heavy works it can be the largest. Nobody’s published percentage is your percentage, and the calculation takes an afternoon.
- Number 2 — cost per cubic metre. Absolute spend tells you nothing; spend per m³ of treated water, trended monthly, is the number that shows dose creep before the invoice does.
- Number 3 — the optimization headroom. Double-digit reductions are routinely claimed by firms selling dosing optimization. Treat any specific figure as a hypothesis to test with jar tests at your own seasonal raw-water extremes, not as a budget line.
- Number 4 — 71 mg/L versus 90–100 mg/L. A peer-reviewed optimization study at a surface-water plant cut the alum dose from a conventional 90–100 mg/L to 71 mg/L with better turbidity and colour removal: more than 30% less chemical and savings above GHS 600,000 a year [1].
- Number 5 — ±5–10% versus ±1–2%. Dosing-pump repeatability spans that range depending on pump type, and on high-cost, low-dose chemicals the gap is worth more than the price difference between pump classes.
- The price line moves on its own. Chemical prices track commodity markets and rise with summer demand, which is why utilities negotiate annual contracts with price caps and keep at least two verified suppliers for critical chemicals.
Number 1 — The Share of OPEX You Actually Control
Start with the denominator, and compute it yourself.
Cost guides for water and wastewater treatment usually place chemicals behind energy and ahead of everything else in the controllable-expense column. The exact share swings enormously with process and region: a utility dosing chlorine, fluoride and a little antiscalant may spend a few percent of its operating budget on chemicals, while a surface-water works running high coagulant doses through a soft season can spend a third or more. Both plants appear in the same trade press under the same headline.
So take twelve months of chemical invoices, twelve months of operating cost, and divide. The exercise takes an afternoon and it is the only version of Number 1 that you can defend in a budget meeting.
What the exercise usually reveals is dose creep. Each operator adds a margin over the last operator’s margin, nobody removes one, and the chemical share drifts upward over years without a single decision that anyone remembers making.
Number 2 — Cost per Cubic Metre: The Denominator That Makes Comparison Possible
Absolute chemical spend is meaningless without a volume denominator.
Normalize spend per cubic metre of treated water, then trend it monthly against production. That single line exposes the two classic failure modes of dosing economics: programmes that overdose at a fixed rate whatever the raw water is doing, and programmes that dose by folklore (“it looked cloudy today”). Both show up as a per-m³ line that climbs while output stays flat.
The absolute figure is plant-specific and not worth benchmarking against a stranger’s plant. The trend is the KPI. So is the comparison across your own lines: two trains making the same product at different doses are telling you something that no supplier will.
Number 3 — The Optimization Headroom: Double Digits, If the Dose Was Wrong to Begin With
The most quoted number in dosing economics is the savings range, and it comes almost entirely from parties who sell optimization. Firms working in DAF and physico-chemical treatment put the achievable operating-expenditure reduction from better coagulant selection, polymer choice and pH correction in the double digits. Industrial audits tell the mirror story, that plants routinely overspend on PAC and PAM before anyone looks.
Both claims are plausible and neither is a promise about your plant. What is reliable is the method:
- Baseline the dose over at least a full seasonal cycle.
- Jar-test at the raw-water extremes, not at a comfortable average.
- Automate only what the jar tests prove.
A first-round optimization that lands well short of the headline percentage is still a good outcome. A first round that hits the headline percentage usually means the baseline was taken during a bad week.
Number 4 — The Proof That Physics Allows It: 90–100 → 71 mg/L
Skeptics reasonably ask whether published savings survive peer review. Some do.
A study at the Weija treatment plant in Ghana combined conventional jar testing with historical operating data from 2002–2016 and 2018 seasonal experiments, then applied response-surface methodology to the coagulation step. Conventional jar tests called for 90–100 mg/L of alum. The optimized model achieved better turbidity and colour removal with 71 mg/L, while also reducing residual aluminium: chemical usage down by more than 30%, cost savings above GHS 600,000 a year, and compliance with WHO guideline values [1].
The generalizable lesson is not the milligrams. Your water is not their water. It is the method: dose requirements were re-derived against measured raw-water quality instead of inherited from historical setpoints. Re-deriving the dose per season, rather than per decade, is where most of the available headroom sits.
Number 5 — The Accuracy That Protects Every Other Number: ±5–10% vs ±1–2%
Every saving above can be quietly re-spent through poor repeatability.
Dosing-system design guidance reports typical repeatability of solenoid diaphragm metering pumps at ±5–10%, against ±1–2% for peristaltic and plunger designs. The economics are asymmetric. On high-cost, low-dose chemicals such as polymer, antiscalant and biocide, a few percent of dose error can exceed the entire purchase-price difference between pump classes.
Ask any supplier for dose repeatability at your actual turndown, not at nameplate. Accuracy is always quoted at the friendly end of the curve.
| Number | What to use it for | Basis | Action |
|---|---|---|---|
| 1. Chemical share of treatment OPEX | Sizing the prize before you negotiate | Your own twelve-month invoices and operating cost | Compute your split; treat any published percentage as an illustration |
| 2. Chemical OPEX per cubic metre | Detecting dose creep and comparing lines | Your own metering and invoices | Trend it monthly against production volume |
| 3. Optimization headroom | Setting expectations for a first-round project | Vendor and consultancy claims, double-digit | Baseline a full season, jar-test at extremes, automate what the tests prove |
| 4. Dose after formal optimization | Benchmarking how far a dose can move | Peer-reviewed plant study, Weija (Ghana) [1] | Re-derive dose per season against measured raw-water quality |
| 5. Dosing pump repeatability | Choosing between pump classes | Dosing system design guides | Upgrade accuracy on low-dose, high-cost chemicals; demand repeatability at your turndown |
The Sixth Factor: Prices Move Whether You Optimize or Not
Optimization controls the dose. Procurement controls the unit price, and the unit price does not stand still.
Utility sourcing guidance notes that treatment-chemical prices track commodity markets and rise with summer demand. The usual defences are annual contracts with price caps or indexed adjustments, inventory sized by economic order quantity so bulk discounts are balanced against storage cost and degradation, and at least two verified suppliers for anything critical, which reduces single-source risk and keeps the negotiation honest.
A practical contract package therefore has three layers: a price mechanism (caps or indexation), a quality anchor (certification to the relevant drinking-water standard plus a certificate of analysis per lot), and a dose anchor (the optimized setpoints from Number 4, so a supplier change never quietly resets your dose).
Closing the Loop: Feedback Beats Guesswork
Numbers 1 to 5 converge on one conclusion that is architectural rather than commercial: dosing cost is a control problem.
A feedback loop with online turbidity, pH and conductivity on the process side, flow-proportional dosing on the reagent side, and a quarterly jar-test recalibration of the setpoints turns dose creep into a visible deviation instead of an invisible habit. It also changes the procurement conversation, because once dose is instrumented every quote can be evaluated against measured normalized consumption rather than against last year’s invoice.
Shanghai ChiMay builds its online water quality analyzers for exactly that role: continuous, integration-ready measurement that feeds dosing control instead of a dashboard nobody prices against. We quote five-year cost, not just unit price, because the unit price of anything, sensor or sulphate, is the smallest number on the page.
Chemical dosing cost management is not a slogan. It is five numbers and a feedback loop. Compute them once and every later procurement decision gets cheaper to justify.
References
- Akkutteh, H.M. and Wiafe, S. — “Optimizing coagulation efficiency in surface water treatment using response surface method”, Journal of Environmental Science and Health, Part A (2025): conventional jar tests 90–100 mg/L alum versus 71 mg/L optimized, chemical usage cut by more than 30%, savings exceeding GHS 600,000 annually, Weija Water Treatment Plant, Ghana. https://doi.org/10.1080/10934529.2025.2594359
About the author: Written by the Shanghai ChiMay Technical Editorial Team — engineers who support procurement and process teams specifying Shanghai ChiMay’s online water quality analyzers and dosing-control instrumentation, with a focus on five-year cost of ownership rather than unit price.
